Mayer Brown, in association with the College of Europe is proud to publish Peter Cimentarov’s thesis entitled “Expanding the “Object Box” and its Perverse Effects. Does EU Competition Law Condemn Innocent Behaviour?” winner of the Mayer Brown Legal Business Prize 2014/2015 for the best thesis concerning the impact of the EU law on business.

Thesis

COLLEGE OF EUROPE BRUGES CAMPUS LAW

Expanding the “Object Box” and its Perverse Effects

Does EU Competition Law Condemn Innocent Behaviour?

Supervisor: Pierre LAROUCHE Thesis presented by Co-supervisor: Mike WALKER Petar CIMENTAROV for the

Degree of Master of European Studies  Option: European Law and Economic Analysis

Academic Year 2013-2014

ii

Statutory Declaration I hereby declare that this thesis has been written by myself without any external

unauthorised help, that it has been neither presented to any institution for evaluation nor

previously published in its entirety or in parts. Any parts, words or ideas, of the thesis,

however limited, and including tables, graphs, maps etc., which are quoted from or based

on other sources, have been acknowledged as such without exception.

Moreover, I have also taken note and accepted the College rules with regard to plagiarism

(Section 4.2 of the College study regulations).

14.970 words

iii

Abstract Background The notion of “anti-competitive object” in the post-modernisation era of competition law has expanded in two ways. First, “new types” of practices have been introduced to the object box. Second, the ECJ has recently suggested that a limited economic analysis may be employed to convert a classic restriction “by effect” into a restriction “by object”. Purpose The purpose of this paper is (i) to examine our present-day understanding of what is an object restriction and (ii) to assess whether a broad interpretation of the notion and its consequences does not lead to a risk of deterring potentially consumer-enhancing behaviour. Analysis First, the paper presents the different methods for identifying an object restriction and brings the abridged competitive analysis approach of the ECJ in Allianz Hungária into focus. Second, it examines the appropriate model for selecting the practices that are suitable for a presumption of harm and do not merit a full-blown effects assessment. Third, the paper analyses whether the contents of the object box, and in particular the recent addition of pay-for-delay arrangements, are justified on grounds of economic theory of harm. Finally, it is argued that the notion of object restrictions has expanded in parallel with post-modernisation developments limiting the possibility for defendants to rebut the presumption of harm. Conclusion It concludes that object restrictions under EU competition law include practices that do not necessarily lead to harm and may contain redeeming virtues. When combined with a significant limitation of the available defence options to rebut the presumption of harm, the broad view of the object box may lead to an approach that is too formalistic, and may therefore result in errors of over-enforcement.

iv

Keywords Article 101 TFEU

Article 101(3) TFEU

De Minimis

Effects-based approach

Information exchanges

Object restrictions

Over-enforcement

Presumption of harm

Pay-for-delay

v

Table of Contents Abstract ............................................................................................................................... iii  

Keywords ............................................................................................................................. iv  

List of Abbreviations .......................................................................................................... vi  

Introduction ......................................................................................................................... 1  

1   What are object restrictions? ....................................................................................... 4  

1.1   Identifying object restrictions ................................................................................... 4  

1.1.1   Textual approach: “l’objet du contrat” .............................................................. 4  

1.1.2   Contextual approach: “legal and economic context” ........................................ 6  

1.1.3   Abridged competitive analysis approach: determining object by looking into

effects ........................................................................................................................... 7  

1.2   Ratio legis of the object/effect dichotomy ............................................................. 11  

1.2.1   Object or effect: alternatives ........................................................................... 11  

1.2.2   Balancing exercise: likelihood of harm vs costs of economic analysis .......... 12  

1.2.3   “Risk offences”: punishable regardless of harmful effects ............................. 14  

1.3   AG Wahl: the voice of clarity ................................................................................ 15  

2   Infatuation with objects: expanding an already large object box .......................... 17  

2.1   Classic object restrictions ....................................................................................... 18  

2.1.1   Agreements hindering parallel trade ............................................................... 18  

2.1.2   Resale price maintenance (RPM) .................................................................... 19  

2.2   Expanding the box beyond classic object restrictions ............................................ 20  

2.2.1   Reverse payment patent settlements (pay-for-delay) ...................................... 20  

2.2.1.1   Traditional position towards patent settlements: effects analysis ............ 20  2.2.1.2   Lundbeck et al.: object restrictions .......................................................... 22  2.2.1.3   Does pay-for-delay fit the object box? ..................................................... 24  

2.2.2   Information sharing ......................................................................................... 27  

3   The perverse effects: relying on presumptions ......................................................... 30  

3.1   Irrebuttable presumption of appreciable anti-competitive effects .......................... 30  

3.1.1   No need to show anti-competitive effects ....................................................... 30  

3.1.2   No possibility to demonstrate absence of anti-competitive effects ................. 31  

3.1.3   No de minimis .................................................................................................. 32  

3.2   Presumption of not satisfying Article 101(3) conditions ....................................... 34  

Conclusion .......................................................................................................................... 37  

Bibliography ...................................................................................................................... 38  

vi

List of Abbreviations ATP Absolute Territorial Protection

CMLR Common Market Law Review

EU European Union

ECHR European Convention on the Protection of Human Rights and Fundamental Freedoms

ECtHR European Court of Human Rights

ECJ European Court of Justice

E.C.R European Court Reports

E.L.Rev. European Law Review

JECLAP Journal of European Competition Law & Practice

NCA National Competition Authority

OECD Organisation for Economic Co-operation and

Development

OJ Official Journal of the European Union

RPM Resale Price Maintenance

TFEU Treaty on the Functioning of the European Union

1

Introduction “Il vaut mieux hasarder de sauver un coupable que de condamner un innocent.” (Voltaire)

Competition law in the past 20 years has embraced economic thinking with a view to

ensuring more accurate results. In the field of Article 102 TFEU, the Commission binds

itself in its Guidance Paper to move away from the previous formalistic approach towards

abuse of dominance and to focus on practices harmful to competition, and ultimately to

consumers, rather than those harmful to competitors.1 To achieve this goal, the Commission

suggests to apply fairly complex price/cost tests in order to verify whether an “as efficient”

competitor will be excluded from the market.2 There are some indications that the Courts

may follow this more economics-based approach.3 In the field of merger control, the main

development is the movement away from a test based on the structure of the market towards

a test based on the expected effects on competition. The new “significant impediment to

effective competition” test4 aims at avoiding both over-enforcement and under-enforcement,

thereby resulting in a more accurate assessment.5

Despite the significant progress made in incorporating economic theory into Article

1016 TFEU7, the approach in the past 10 years to practices having an anti-competitive object,

or object restrictions, raises serious concern as regards the appropriate enforcement level in

competition law. As Voltaire put it centuries ago, “it is better to risk saving a guilty man

than to condemn an innocent one”8. His view is relevant to present-day competition law in

two ways. First, this liberal vision and preference for under-enforcement 9 to over-

1 Commission Communication of 24 February 2009, Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings, O.J. C 45, para 19. 2 Ibidem, para 25 and 26. 3 See Case C-209/10, Post Danmark A/S v Konkurrencera det (not yet reported), judgment of 27.03.2012. 4 Council Regulation (EC) No 139/2004 of 20 January 2004, on the control of concentrations between undertakings, [2004] O.J. L 24, Article 2(3). 5 Under the new test, dominance is neither “sufficient” nor “necessary” for a concentration to be prohibited. See L-H RO LLER & M. DE LA MANO, “The Impact of the New Substantive Test in European merger Control” (2006) 2(1) European Competition Journal, p.22. 6 Reference is made in particular to the regulations, exempting categories of agreements, both vertical and horizontal, from the application of Article 101. In line with economic theory, given the parties’ lack of market power and the absence of clauses capable of seriously restricting competition, the agreements are regarded as not having the potential to produce appreciable anti-competitive effect. 7 All references to Article 101 relate to Article 101 of the Treaty on the Functioning of the European Union. 8 Free translation from the quote in French at the top of the page. 9 I.e. allowing practices which do result in anti-competitive harm. Also called, false negative or Type 2 errors.

2

enforcement10 is the “prevailing view” in competition law nowadays, as the latter is likely to

“chill pro-competitive activity and stunt innovation”.11 Second, now that the criminal nature

of the Commission’s and NCA’s sanctioning powers has been established 12 , over-

enforcement is all the more of a concern, given that object restrictions usually result in the

imposition of a hefty fine. The purpose of this paper is to examine whether the current

interpretation in EU competition law of what constitutes an object offence and what its

consequences are, does not lead to deterring and condemning “innocent” behaviour.

It is submitted in this paper that five post-modernisation13 developments in relation to

object restrictions have limited the role of economics in Article 101, ultimately leading to a

higher risk of over-enforcement:

1. The extension of the object box14 with new types of practices, such as some forms of reverse payment patent settlements and information exchanges.

2. The abridged competitive analysis approach for identifying object restrictions, proposed by the ECJ in Allianz Hungária15.

3. The perception of object restrictions as “risk offences”, as suggested by AG Kokott in her Opinion in T-Mobile16, and the resulting irrebuttable presumption of harm under Article 101(1).

4. The judgement of the ECJ in Expedia 17 leading to the exclusion of de minimis arguments in the context of object restrictions.

5. The creation of a strong presumption that object restrictions do not satisfy the conditions for Article 101(3) and the lack of guidance on how to rebut the presumption.

Whereas the first two developments relate to expanding our understanding of object

restrictions, the last three concern the perverse effects for defendants flowing from a

classification of an agreement as an object restriction.

10 I.e. prohibiting practices which do not result in anti-competitive harm. Also called, false positive or Type 1 errors. 11 JONES & SUFRIN, EU Competition Law, 5th ed., Oxford University Press, 2014, p.57-58.. 12 See ECtHR, Menarini Diagnostics S.R.L. v. Italy, judgment of 27 September 2011, No 43509/08; Opinion of AG Sharpston in Case C-272/09 P, KME Germany AG v. Comission, delivered on 10 February 2011, para 64. 13 The modernisation of competition policy started around 1996 and involved not only a movement from form-based to effects-based approach in substantive competition law, but also institutional and procedural developments. See JONES, “Left Behind by Modernisation? Restrictions by Object under Article 101(1)” (2010) 6(3) European Competition Journal, p.649-656; GERARD, “The Effects-Based Approach Under Article 101 TFEU and its Paradoxes: Modernisation at War with Itself?” in BOURGEOIS and WAELBROECK (eds.), Ten Years of Effects-Based Approach in EU Competition Law, Bruylant, 2012, p.18-22. 14 In this paper the word “object box” is used to describe all restrictions regarded as having an anti-competitive object under EU competition law. See WHISH, infra note 27. 15 Case C-32/11, Allianz Hungária Biztosító Zrt and Others v Gazdasági Versenyhivatal (not yet reported), judgment of 14.03.2013. 16 Opinion of AG Kokott in Case C-8/08, T-Mobile, delivered on 19 February 2008. 17 Case C-226/11, Expedia Inc. v. Autorité de la Concurrence (not yet reported), judgment of 13.12.2012, para 37.

3

Chapter 1 examines what different methods are employed to identify an object

restriction (1.1) and what model should be used, from a policy perspective, to select the

practices typically regarded as object restrictions (1.2). Chapter 2 provides a critical look at

the current contents of the object box and focuses in some more detail on pay-for-delay

arrangements (2.2.1). Chapter 3 analyses the impact of recent post-modernisation

developments on the possibility for an object restriction to avoid infringing Article 101.

4

1 What are object restrictions?

1.1 Identifying object restrictions

Some authors argue that recognizing an object restriction is supposed to be an easy

task: every legal counsel, not versed in economics of competition, should be able to detect a

clause anti-competitive by object without the need to look further into the specific market

circumstances surrounding the agreement (1.1.1). 18 Other authors submit that such

formalistic reasoning would not reflect the case-law of the European Courts: one cannot

simply limit object restrictions to a list of practices, but has to perform a case-by-case

assessment of each practice in the broader legal and economic context within which it takes

place (1.1.2).19 In Allianz Hungária20, the ECJ comes up with a third method, which

arguably stretches the limits of the contextual approach: an object restriction can be

identified by a “quick look” into the likely effects of the agreement (1.1.3).

1.1.1 Textual approach: “l’objet du contrat”

“I know it when I see it” (Justice Potter Stewart in Jacobellis v. Ohio21 regarding possible obscenity)

The famous quote by Justice Potter Stewart reflects perfectly the idea that some

offences are obvious to the naked eye. If you do not see it, then it is not there. This reasoning

is similar to the textual approach towards object restrictions. It suggests that in order to

determine whether an agreement contains such a restriction, one should look only at the

terms of the agreement and should not look further than the “four angles of the contract”22.

The textual reading of object offences is arguably influenced by civil law legal

traditions. The French notion of “l’objet du contrat” means the legal transaction

contemplated by the parties, as determined from all the rights and obligations which the

contract creates, at the time of the conclusion of the contract.23 The objective of an

18 NAGY, “The Distinction between Anti-competitive Object and Effect in Allianz: The End of Coherence in Competition Analysis?”, (2013) 36(4) World Competition, p.555. 19 KING, “The Object Box: Law, Policy or Myth?”, (2011) 7(2) European Competition Journal, p.270. 20 Allianz Hungária, supra note 15. 21 US Supreme Court, Jacobellis v. Ohio, (1964), 378 U.S. 184. 22 NAGY, supra note 18, p.544. 23 The provision of the French Civil Code dealing with the object of the contract is Article 1126. There are two readings of this provision in the French legal literature (See PELLETIER, La caducité des actes juridiques en droit privé français, Paris, Harmattan, 2004, p.43). For example, “l’objet du contrat” can mean either the house, which is being sold, or the sale itself. The latter interpretation is more clearly defined in Article 1412 of the Civil Code of Quebec, from where the quote in the text originates: “l’objet du contrat est l’opération juridique

5

agreement may be, for example, to grant a license or to agree not to compete in certain

geographical areas. This should be determined on the basis of the clauses contained in the

contract. The main function of identifying “l’objet du contrat” in legal systems, wearing the

Napoleonic stamp, is to examine whether it does not violate public order considerations.24

For instance, agreements having as main objective the sale of a person or the formation of a

cartel both go against public order, and are therefore null and void.

This rather formalistic reasoning, which does not take into account the broader

economic context of the agreement, has left its, albeit rather limited, mark on EU

competition law. The Commission states in its Article 101(3) Guidelines that “restrictions of

competition by object are those that by their very nature have the potential of restricting

competition”.25 Furthermore, the distinction between object and effect, as understood by the

supporters of the more formalistic textual approach, is best exemplified by the General Court

in European Night Services. The Court states that in order to assess a practice under Article

101(1), "account should be taken of the actual conditions in which it functions, in particular

the economic context in which the undertakings operate, the products or services covered by

the agreement and the actual structure of the market concerned, unless it is an agreement

containing obvious restrictions of competition such as price-fixing, market-sharing or the

control of outlets”.26 In other words, there are certain categories of practices which are under

any circumstances presumed to restrict competition under Article 101(1).

According to Whish, one can think of competition law infringements in terms of

boxes: the object box and the effect box. 27 The former is a small box containing only

restrictions that are “so clearly inimical to the objectives of the EU that they can be permitted

only where they can be shown to satisfy the requirements of Article 101(3)”.28 Based on the

case-law of the European Courts and the decisional practice of the Commission, Whish

defines the contents of its object box as follows: “(i) horizontal agreements to fix prices, to

exchange information that reduces uncertainty about future behaviour, to share markets, to

limit output, including the removal of excess capacity, to limit sales and for collective

envisagée par les parties au moment de sa conclusion, telle qu’elle ressort de l’ensemble des droits et obligations que le contrat fait naître”. 24 Article 6 of the French Civil Code. 25 Commission Communication of 27 April 2004, regarding Guidelines on the Application of Article 81(3), [2004] O.J. C101/97 (hereinafter “the Article 101(3) Guidelines”), para 21. See also Case C-209/07, Competition Authority v. Beef Industry Development Society and Barry Brothers, [2008] ECR I-8637, para 17. 26 See Joined Cases T-374, 375, 384 & 388/94, European Night Services v. Commission, [1998] ECR II-3141, para 136. 27 WHISH, Competition Law, 7th ed., Oxford University Press, 2012, p.120. 28 Ibidem, p.121.

6

exclusive dealing and (ii) vertical agreements to impose fixed or minimum resale prices and

to impose export bans”.29 This suggests that these practices have the object to restrict

competition, regardless of the legal and economic context within which they take place. The

key question is therefore whether an agreement can be qualified as one of the agreements

found in the object box.

There are clear advantages of using this rather simplistic approach of thinking in

terms of categories of infringements. The obvious one is legal certainty. Legal counsels can

benefit from a list of “no-go” practices. In addition, undertakings will have the certainty that

any practice outside of the object box will be given a full-fledged economic assessment.

1.1.2 Contextual approach: “legal and economic context”

“We need to remember: restrictions by object are serious - but not necessarily obvious.” (A.Italianer)30

The second approach to identifying an object restriction, albeit more sophisticated, is

the one that prevailed in the case-law. It claims that rather than assessing agreements in a

“vacuum”, account should be taken of the legal and economic context within which the

agreement takes place. 31 This means that two contracts containing the same wording can be

either qualified as restrictions having an anti-competitive object or assessed on their effects,

depending on factors external to the contract. Such factors include, for example, the type of

industry, the existence of specific IP rights32 or of documentary evidence showing intent33.

The European Courts have embraced the contextual analysis in the majority of their

judgements.34 In addition, the Commission in its Article 101(3) Guidelines specifies that “an

examination of the facts underlying the agreement and the specific circumstances in which it

operates may be required before it can be concluded whether a particular restriction 29 Ibidem., p.124. The same restrictions grosso modo are enlisted in the definition of a hardcore cartel in OECD Council 1998, “Recommendation of the Council concerning Effective Action against Hard Core Cartels”, adopted by the Council at its 921st session on 25 March 1998 and reprinted in OECD Report, Fighting Hard-Core Cartels: Harm, Effective Sanctions and Leniency Programmes, OECD Publications, Paris, 2002, pp. 105-107 and in the list of object restrictions included in the Article 101(3) Guidelines, supra note 25, para 23. 30 Speech of ITALIANER, “Competitor Agreements under EU Competition Law”, 40th Annual Conference on International Antitrust Law and Policy at Fordham, New York, 26 September 2013. Available at: http://ec.europa.eu/competition/speeches/text/sp2013_07_en.pdf, April 20, 2014. 31 ITALIANER, supra note 30, p.5. 32 See Case 262/81, Coditel v Ciné Vog Films, [1982] ECR 3381. 33 Proof of subjective intent is not a necessary condition for a finding of an object restriction, see Article 101(3) Guidelines, para 22. Odudu goes even further and argues that although subjective intent is not a conditio sine qua non for the finding of “object”, it is a sufficient condition, see ODUDU, “Interpreting Article 81(1): Object as subjective intention”, (2001) 26(1) E.L.Rev., p.60-75. 34 See Joined Cases C-501/06, C-513/06, C-515/06 and C-519/06, GlaxoSmithKline v. Commission, [2009] ECR. I-9291, para 58; BIDS, supra note 25, paras 16 and 21.

7

constitutes a restriction of competition by object” 35 . The assessment of external

circumstances is, however, not supposed to go as far as to examine the effects of the

agreement on the market. Otherwise, that would lead to an odd situation where

demonstrating some anti-competitive effects is required in order to establish an object

restriction.36

The contextual analysis is a more refined approach to determining what is an object

restriction. It allows for a tailored case-by-case assessment of whether the agreement

deserves wearing the object “stigma”. Nonetheless, it inevitably leads to more legal

uncertainty and can act as a “double-edged sword”. On the one hand, looking outside the

boundaries of the agreement may serve to pull some typically object restrictions out of the

object box, and in some cases even out of the reach of Article 101(1). Both Coditel II37 and

Erauw-Jacquéry38 concerned ATP clauses which were found to fall outside Article 101. The

ECJ looked beyond the “four angles of the contract” and placed the agreements in their

economic context, paying particular attention to the specific nature of the property rights

involved 39 and the necessity to protect investments. 40 On the other hand, external

circumstances can place a typically effects restriction within the scope of the object box.

Chapter 1.1.3 presents the example of Allianz Hungária41.

1.1.3 Abridged competitive analysis approach: determining object by looking into effects

“[T]he line between restrictions by object and those by effect is not always bright. Reading its more recent rulings one may wonder whether the Court, whilst finding a restriction by object, may not have gone rather far towards analysing the effects of the agreement…”42 (A. Italianer on Allianz Hungária)

The abridged competitive analysis 43 approach is an extended version of the

contextual approach. It implies that in order to determine whether a restriction has an anti-

35 Supra note 25, para 22. 36 See infra Chapter 1.1.3. 37 Coditel II, supra note 32. 38 Case 27/87, Erauw-Jacquéry v. La Hesbignonne, [1988] ECR 1919, para 11. 39 While Erauw-Jacquéry involved protection granted to a basic seed licensee, Coditel II involved an exclusive right to the public performance of a film. 40 Erauw-Jacquéry and Coditel II illustrate well the application of the “ancillary restraints” doctrine in EU competition law. For further reading, see JONES & SUFRIN, supra note 11, p.242-246. 41 Supra note 15. 42 ITALIANER, supra note 30, p.6. 43 The term originates from Case T-168/01, GlaxoSmithKline Services v Commission, [2006] ECR II-2969, paras 119-120; See BAILEY, “Restrictions of Competition by Object under Article 101 TFEU”, (2012) 49 CMLR, p.585-586.

8

competitive object, a truncated analysis of the competitive structure of the market, including

an examination of market power, may be conducted. Hence, this method falls short of

performing a full-fledged assessment of anti-competitive effects.

A recent preliminary ruling judgement of the ECJ, Allianz Hungária44, illustrates

well the technique of the abridged competitive analysis. The case raised concerns amongst

practitioners that the EU and the Member States are identifying object restrictions by

conducting a limited analysis of effects, thereby eliminating any meaningful distinction

between the two.45 Allianz Hungária concerned vertical agreements46 between insurance

companies, Allianz and Generali, and car repairers or an association of car repairers. The car

repairers wore two hats: they acted as repairers for consumers and as insurance brokers for

insurance companies. The provisions of the agreements stipulated that the hourly repair

charge paid by the insurance company to the repairer was a function of the number and

percentage of insurance policies purchased from the insurance company. In other words, the

remuneration of car repairers increased with the number of insurances sold to customers.

This created an incentive for car repairers to convince customers to buy an insurance policy

from Allianz or Generali. The Hungarian competition authority took the “shortcut” option: it

classified the agreement as having an anti-competitive object and did not make the effort to

look into effects.47 The question referred to the ECJ was whether such an agreement could be

qualified as an object restriction.48

AG Cruz Villalón followed a rather traditional reasoning, in line with previous case-

law, to conclude that such an agreement deserved a full-fledged effects analysis.49 Firstly, he

stated that competition law did not expressly prohibit vertical clauses which produce

incentives for the distributor to sell more units from the supplier’s product, even if the clause

was “intended to increase an undertaking’s own sales at the expense of those of its

competitors”.50 After all, even single-branding obligations51, if limited in time, are tolerated

44 Supra note 15. 45 See HARRISON, “The Allianz Hungária case - The ECJ's judgment could have ugly consequences”, (2013) 12(6) Competition Law Insight, p.10-12; NAGY, supra note 18. 46 The case had a horizontal aspect as well, which is not discussed here. 47 NAGY, supra note 18, p.556. 48 Allianz Hungária, supra note 15, para 16 49 Opinion of AG Cruz Villalón in Case C-32/11, Allianz Hungária, delivered on 25 October 2012, paras 61-81 50 Ibidem, para 77. 51 The agreement at hand in Allianz Hungária was not even a de jure non-compete obligation, but a clause which might de facto have the same effect as a non-compete clause.

9

by EU competition law.52 Secondly, the AG referred to earlier case-law, where the Courts

examined non-compete clauses on account of their effects.53

Unlike AG Cruz Villalón, the ECJ did not exclude that the agreement at hand,

“following a concrete and individual examination of the wording and aim of those

agreements and of the economic and legal context of which they form part”, could be

classified as an object restriction.54 The ECJ gave guidance to the referring court as to the

elements which should be considered when determining whether the specific clause had an

anti-competitive object. Two of them appear novel and seem to deviate from previous case-

law on similar vertical restraints.

First, the ECJ submits that Hungarian consumer protection laws, requiring that car

repairers acting as insurance brokers be independent from insurance companies, is a relevant

criteria for identifying an object restriction.55 It looks like the ECJ provides for a possibility

to correct the malfunction of another field of the law56, consumer protection, by the

application of competition law.57

Second, the ECJ holds that an object restriction can be found where, “having regard

to the economic context”, and in particular “the structure of the market, the existence of

alternative distribution channels and their respective importance and the market power of

the companies concerned”, “it is likely that […] competition on the market would be

eliminated or seriously weakened”.58 This reasoning is problematic in two ways. From a

normative point of view, it creates confusion as to where object analysis stops and effects

analysis begins. It is striking that the ECJ establishes a test for determining object

restrictions, which is based on a limited analysis of the competitive structure, in order to

conclude that a more in-depth analysis is unnecessary. In fact, the criteria mentioned by the

ECJ in paragraph 48 are almost identical to the relevant criteria employed to establish

52 Opinion of AG Cruz Villalón in Case C-32/11, Allianz Hungária, supra note 49, para 77 53 Ibidem, para 80; See Case C-234/89, Delimitis [1991] ECR I-935, paras 13-15. 54 Allianz Hungária, supra note 15, para 51. 55 Allianz Hungária, supra note 15, para 47. 56 A similar idea can also be found in another recent case, see Case C-457/10 P, AstraZeneca AB and AstraZeneca plc v. European Commission (not yet reported), judgment of 6.12.2012, where competition law provided a remedy for the malfunction of the patent system. 57 See NAGY, supra note, p.561, who states that “perverted cases normally entail perverted judgments”; VEDDER, “Allianz and the Object-Effect Dichotomy in Article 101(1) TFEU: A Practical Solution Meets Not So Practical Competition Law”, European Law Blog, 8 April 2013. Available at: http://europeanlawblog.eu/?p=2147#sthash.6FYzqXCT.dpuf, who regards the Court’s reasoning as a “spectacular misconception of competition law”. 58 Allianz Hungária, supra note 15, para 48.

10

foreclosure in classic effects cases59, such as Delimitis60 and Brasserie de Haecht61. Hence, it

is not entirely clear to what extent the abridged competitive analysis, as proposed by the ECJ

in Allianz Hungária, differs form the full-fledged analysis, applied by the Courts in Delimitis

and Brasserie de Haecht. From a practical point of view, the judgment appears to authorize

the Commission and the NCAs to present as much economic evidence as they have in

possession, or are willing to collect, so as to classify a practice as an object restriction. Such

a policy will not only deprive undertakings of their legitimate expectations to have their

agreements assessed under a full-blown economic analysis, but will also significantly reduce

the odds of a successful defence62. Hence, the conversion of a typically effects restriction

into an object one, is more likely to lead to an incorrect result. When the external

circumstances relevant for the abridged assessment are not such as to guarantee that

consumer harm would inevitably occur, a comprehensive analysis of effects is more

appropriate in order to avoid a Type 1 error - the risk of over-enforcement.

59 The two cases concerned exclusive purchasing provisions for the supply of beer in vertical agreements between breweries and pubs. 60 Supra note 53. 61 Joined Cases T-49-51/02, Brasserie Nationale v. Commission, [2005] ECR II-3033. 62 Object offences face an irrebuttable presumption of appreciable harm under Article 101(1) and a presumption of not satisfying the conditions for Article 101(3). See infra Chapter 3.

11

1.2 Ratio legis of the object/effect dichotomy

Having examined the three methods employed to identify an object restriction, we

now turn to analysing the very rationale behind the existence of the object/effect divide. This

Chapter explores the following questions: what is the relationship between object and effect

(1.2.1); what is the appropriate model to select which practices should be regarded as object

restrictions (1.2.2); are object restrictions sanctioned even when they fail to produce anti-

competitive effects (1.2.3).

1.2.1 Object or effect: alternatives

“The concept of restriction of competition is an economic one, and as a general proposition economic analysis is needed to determine whether an agreement could have an anti-competitive effect.”63 (Whish)

Article 101(1) prohibits agreements which “have as their object or effect” restriction

of competition. The conjunction “or” suggests that effect and object are not cumulative

requirements, but alternatives. The Courts have confirmed this view in several cases.64

The mechanism of Article 101(1) is as follows. If it is demonstrated that the

agreement has an anti-competitive object, there is no need to look at its effects on the

market. It is presumed to have anti-competitive effects. However, if no anti-competitive

object is proven, an analysis of the effects of the agreement must take place. To perform this

exercise, it is necessary to establish a counterfactual. The ECJ stated in John Deere that “in

order to determine whether an agreement is to be considered to be prohibited by reason [of]

its effects, the competition in question should be assessed within the actual context in which

it would occur in the absence of the agreement in dispute”65.

The Commission often reminds that proof of an infringement of Article 101(1) is not

“the end of story”.66 As a matter of principle67, in both scenarios, whether a restriction by

object or by effect is established, undertakings can revert to Article 101(3) in an attempt to

justify the agreement on account of efficiencies offsetting the negative anti-competitive

effects.

63 WHISH, supra note 27, p.117 64 Case C-56/65 Société Technique Minière (L.T.M.) v Maschinenbau Ulm GmbH (M.B.U.) [1966] ECR 235, para 249. 65 Case C-7/95P, John Deere Ltd v. Commission, [1998] ECR I-3111, para 76. 66 ITALIANER, supra note 30, p.7. 67 It is argued in Chapter 3.2 that, in practice, it is very difficult for undertakings to rely on Article 101(3).

12

1.2.2 Balancing exercise: likelihood of harm vs costs of economic analysis

“Restrictions of competition by object … have such a high potential of negative effects on competition that it is unnecessary for the purposes of applying Article [101] (1) to demonstrate any actual effects on the market.” (Article 101(3) Guidelines68)

Thinking in terms of object restrictions is an imperfect, but a sensible policy choice.

Although the object approach is more likely to lead to errors than an effects-based approach,

it produces significant benefits relating to legal certainty and reasonable use of resources.69

First, it informs undertakings of the type of practices which are presumed anti-competitive

and will not be given a full-fledged effects analysis. Given the criminal nature of the fines70,

it is of utmost importance that an economic operator is fully aware that a certain practice

constitutes a criminal offence71. Second, it reduces enforcement costs, as it avoids complex

and costly economic analysis.

Reliance on a presumption of anti-competitive effects as an essential part of antitrust

policy is not a typically European phenomenon. In Northern Pacific Railway, the US

Supreme Court stated that “there are certain agreements or practices which because of their

pernicious effect on competition and lack of any redeeming virtue are conclusively presumed

to be unreasonable and therefore illegal without elaborate inquiry as to the precise harm

they have caused or the business excuse for their use.”72 The per se approach to antitrust

offences in the US is similar in design, but not identical to the EU object approach. While

the presumption of illegality for US per se restrictions is irrebuttable, EU object restrictions

can, in principle, be justified under Article 101(3). Given their stricter nature, per se

restrictions in the US form a smaller box than object restrictions in the EU.73

Having shown the benefits and the worldwide use of a system based on a category of

practices presumed illegal, we now turn to exploring the criteria used for selecting the

practices that warrant a presumption. In the words of the Commission, the presumption of

harm for object restrictions in the EU is justified by their “serious nature”, and by past

“experience showing that restrictions of competition by object are likely to produce negative

effects on the market and jeopardize the objectives pursued by the Community competition 68 Supra note 25, para 21. 69 Opinion of AG Kokott in T-Mobile, supra note 16, para 43. 70 See Menarini and Opinion of AG Sharpson in KME, supra note 12. 71 The idea that criminal provisions should be sufficiently clear is part of the “nullum crimen, nulla poena sine lege” principle enshrined in Article 7 ECHR. 72 US Supreme Court, Northern Pacific Railway Co. v. United States, 1958 356 U.S. 5. 73 For example, RPM is treated under rule of reason in the US and as an object restriction in the EU. See US Supreme Court, Leegin Creative Leather Products Inc v PSKS, Inc, (2007) 551 US 877.

13

rules”.74 The reference to “experience” should be broadly interpreted so as to include not

only previous EU Court an Commission cases, but also “experience from another

jurisdiction” and “insights of industrial organisation”.75

As a rule of thumb, a good model to determine whether a practice should be

classified as an object restriction is as follows. If, on account of past experience, it is

demonstrated that a practice always (or almost always) leads to a not insignificant harm to

consumers, the practice should be an object offence. Cartels are a good example of practices

suitable for the object box because past experience and theory of harm show with sufficient

certainty that cartels lead to consumer harm.76 On the other hand, if the practice does not

always (or almost always) lead to consumer harm and an economic analysis is fairly easy to

conduct, then it should fall outside the object box. These two scenarios are rather clear-cut.

The problems arise where the behaviour is not always (or almost always) anti-

competitive, but the assessment of effects is complicated, costly and time-consuming.

Zenger & Walker argue that theory of harm related to practices, such as restrictions of

parallel trade, resale price maintenance and multi-lateral interchange fees in payment card

markets, does not demonstrate that they are almost always harmful.77 Theory of harm related

to some new types of infringements, such as reverse payment patent settlements, does not

show conclusive anti-competitive effects either.78 Classifying these practices as object

restrictions in the absence of a robust theory of harm can possibly lead to prohibiting

behaviour which is not necessarily anti-competitive – type 1 error. Such a policy would be

acceptable only if the likelihood of harm caused by the behaviour is typically very high and

close to “always or almost always” and there are strong arguments against conducting an

economic analysis in a particular case.

But, what are the main objections against performing a full-blown economic

assessment in every case where anti-competitive effects are not certain? First of all,

assessment of effects does not always guarantee correct results. This could be due to the

complexity of the exercise, the different possible approaches or the lack of data. Second,

74 Article 101(3) Guidelines, supra note 25, para 21. 75 BAILEY, supra note 43, p.564-565. 76 ZENGER & WALKER, “Theories of Harm in European Competition Law: A Progress Report” in BOURGEOIS and WAELBROECK (eds.), Ten Years of Effects-Based Approach in EU Competition Law, Bruylant, 2012, p.195-196. 77 Ibidem, p.196-202. 78 See infra Chapter 2.2.1.3.

14

sophisticated economic analysis can be very time-consuming79. This is of particular concern

where a timely solution to the anti-competitive practices is of utmost importance.80 On that

account, in some cases there are good policy reasons to avoid complex and time-consuming

economic analysis, where the risk of getting it wrong by using a presumption remains low.

1.2.3 “Risk offences”: punishable regardless of harmful effects

“In a way, it is a bit like speed limits. It is presumed – based on experience – that driving too fast is dangerous, and it is therefore prohibited. That prohibition applies irrespective of whether someone drove safely above the speed limit, whether the offence actually resulted in an accident or whether someone had other, perhaps laudable motives for driving too fast.”81 (A. Italianer)

Object restrictions are more than a shortcut for identifying anti-competitive harm.

Currently, they are regarded as serious violations of the law which should be punished even

if they do not harm consumers in a specific case. To support this view, Mr Italianer recently

compared them to speed limit violations.82 Previously, AG Kokott in her Opinion in T-

Mobile expressed the view that object restrictions are similar to “risk offences” in criminal

law, such as driving under the influence.83 Offenders receive a criminal or administrative

sanction regardless of whether any damage was caused.

The distinction between the two rationales is not purely theoretical and has important

consequences. Under the more economic rationale, which identifies, based on theory of harm

and past experience, practices that always or almost always result in damage to consumers,

the parties should be allowed to present evidence showing that in the specific case their

behavior did not cause harm. In other words, the presumption of harm under Article 101(1)

should be rebuttable. However, this is not the case under the “risk offences” approach

applied by the Commission and the Courts. In practical terms, it is not a valid argument for a

cartel participant to demonstrate that an agreement to raise prices did not produce anti-

competitive effects because, for example, it was not implemented yet or the participants

79 It took the Commission approximately nine years from AMD’s complaint to adopting a decision in Commission Decision of 13.05.2009, COMP/37.990, Intel, OJ C227/13. Five years following the appeal, we are still awaiting the judgment of the General Court. 80 This is the case in fast-moving technology markets and industries with network effects. 81 ITALIANER, supra note 30, p.5. 82 Ibidem. 83 Supra note 16, para 47.

15

constantly cheated by undercutting each other’s prices. AG Kokott in her Opinion in T-

Mobile confirms the impossibility to rebut the presumption of harm under Article 101(1).84

The “risk offences” rationale takes us one step away from a purely economic

approach and one step closer to a criminal enforcement model. Sanctioning behavior

irrespective of whether it caused, or is capable of causing anti-competitive effects, is a sound

policy when applied to cartels. Since they are often secretive and are capable of causing

great damage, cartel-like behavior can be placed in the criminal realm. Given the difficulty

of detecting collusive practices, the need to increase deterrence and the absence of any

redeeming virtues, it makes sense to send a clear signal to undertakings that such behavior

will not be tolerated. The same cannot be said with respect to other object restrictions, such

as resale price maintenance, restrictions of parallel trade, cooperation agreements containing

hardcore restraints, multi-lateral interchange fees and reverse payment patent settlements.

Not only are such practices usually contractual, and therefore not difficult to detect, but they

are also capable of being welfare-enhancing. Thus, in such cases alleged infringers should be

given the chance to demonstrate that their agreements did not have anti-competitive effects.

1.3 AG Wahl: the voice of clarity

The very recent Opinion of AG Wahl in Groupement de cartes bancaires85 sums up

well the appropriate approach to object restrictions and allows us to draw conclusions for

Chapter 1.

As regards the correct method for identifying object restrictions, he expresses a

strong preference for the contextual approach.86 To support his view, he gives the example of

a price-fixing agreement between two companies having a tiny market share.87 Even if such

an agreement, in principle, leads to considerable harm to competition, this is not the case if

the market shares of the parties - the economic context - are extremely low. On the other

side, he recognizes that the contextual approach should know its limits. He argues that it is of

utmost importance to create a clear distinction between, on the one hand, analyzing an

agreement within its economic context in order to determine whether it is an object

84 Ibidem, para 45. 85 Opinion of AG Wahl in Case C-67/13 P, Groupement de cartes bancaires (CB) contre Commission européenne, delivered on 27 March 2014. 86 Ibidem, para 41. 87 Ibidem, para 42.

16

restriction and, on the other hand, conducting an effects assessment88. The contextual

analysis can only be applied to confirm or neutralize the terms of the agreement.89 It should

not go as far as to look at potential anti-competitive effects. It would be inappropriate to seek

to qualify the agreement as having an anti-competitive object by conducting a “quick look”

analysis of effects, so as to compensate for the absence of a “genuine” object restriction.90

As regards the appropriate criteria for deciding which practices should be treated as

“by object” and which practices deserve a full-fledged economic analysis, AG Wahl

reasoned as follows. First, he recognizes that a formalistic approach relying on object

restrictions is not unproblematic.91 Then, he proposes to restrict the notion of object

restrictions only to that behavior whose harmful nature is uncontested, based on experience

and economic science.92 Hence, agreements which either show ambivalent effects on the

market or lead to restrictive effects, which are necessary for achieving a non-restrictive main

objective of the agreement, should not fall in the object box.93 To sum up his understanding

of object restrictions, the irrebuttable presumption of harm should not apply (i) where no

past experience, based on economic theory of harm, shows that the agreement will inevitably

lead to harmful effects or (ii) where the agreement contains a restrictive clause which is

necessary for the main, neutral or positive towards competition, objective of the agreement

to take place. Chapter 2 examines whether the European Courts and Commission follow this

sensible approach.

88 Ibidem, para 44. 89 Ibidem, para 44. 90 Ibidem, para 44; For his criticism of Allianz Hungária, see paras 47-52. 91 Ibidem, para 54 92 Ibidem, para 56: “Ne devraient donc être considérés comme restrictifs de concurrence par objet que les comportements dont le caractère nocif est, au vu de l’expérience acquise et de la science économique, avéré et facilement décelable…” 93 Ibidem, para 56: “…et non les accords qui, au vu du contexte dans lequel ils s’insèrent, présentent des effets sur le marché ou qui sont porteurs d’effets restrictifs accessoires nécessaires à la poursuite d’un objectif principal non restrictif de concurrence”

17

2 Infatuation with objects: expanding an already large object box

There is an ongoing tendency by both the European Commission and the European

Courts towards finding object restriction. In the period between 2000 and 2011 the

Commission issued, excluding cartels, 18 infringement decisions, 17 out of which involved

object restrictions and only one case that was a restriction by effect.94 At the level of the

ECJ, it is barely possible to find an Article 101 judgment which does not conclude that the

practice at hand is an object restriction.95

The Courts and the Commission are pushing the boundaries of the object box, in

some cases beyond what is advisable on grounds of competition economics. On the one

hand, recently, new types of infringements96 have been introduced to the list of object

restrictions. This Chapter examines the suitability of applying a presumption of harm to

some types of information sharing (2.2.2), and more in detail, to reverse payment patent

settlements (2.2.1)97. On the other hand, it is not only the enlargement of the object box that

is questionable, but also some of the vertical agreements that have been there for years. This

Chapter touches briefly upon some of the arguments against applying a presumption of harm

to agreements restricting parallel trade (2.1.1) and to resale price maintenance clauses

(2.1.2). It is submitted that, considering the benchmark established in Chapter 1 for

determining when a restriction should be an object one, namely (i) a conclusive theory of

harm and (ii) absence of a non-restrictive main legitimate objective which would not be

achieved without the restrictive clause, it is questionable whether the four practices deserve

their place in the object box.

94 GERARD, supra note 13, p.38. 95 Ibidem, p.38. 96 New types of infringements, as opposed to classic object restrictions, are those that are not mentioned in the list of the OECD and the Article 101(3) Guidelines, see supra note 29. 97 Reverse payment patent settlements are a highly debated topic at that moment and provide for a good case study of a new addition to the list of object restrictions. Therefore, the Commission’s policy towards this practice will be examined more extensively.

18

2.1 Classic object restrictions

2.1.1 Agreements hindering parallel trade

“Economic evidence shows that it is impossible to state either that all territorial restraints by firms which lack market power are efficiency-enhancing (as some Chicago School lawyers would contend) nor that they are all inefficient (as Consten and Grundig suggests)”98 (G. Monti)

Agreements aimed at limiting parallel trade are traditionally regarded as object

restrictions.99 Examples include export bans or other mechanisms, such as quotas and dual

pricing. Such agreements that restrict the territory to which the distributor may supply, are

“blacklisted” under Article 4(b) of the Vertical Block Exemption Regulation (“VBER”).100

Economic theory states unequivocally that restrictions of parallel trade do not

necessarily have anti-competitive effects.101 In many cases their main objective is to allow

companies to set different prices in different Member States. Such price discrimination can

have welfare-enhancing effects, as it provides firms with “the ability to locally react to the

specific competitive conditions prevailing in different geographic regions”102.

Nevertheless, the ECJ in GSK103 persisted in classifying limitations on parallel trade

as object restrictions. The case concerned an indirect way to restrict parallel trade, namely

via a dual pricing policy. Spanish wholesalers were charged higher prices if they were to sell

the drugs outside of Spain than if they were to sell them in Spain. Interestingly, the General

Court narrowed down the possibility for the Commission to qualify a restriction of parallel

trade as an object restriction. It required that, in order for such a clause to constitute an

object restriction, detriment to the final consumer be proven on the basis of an abridged

economic analysis, an in-depth analysis of the legal and economic context of the

agreement.104 The ECJ overturned the General Court and ruled that no condition of proving

consumer harm is required.105 Although the test proposed by the General Court would have

98 MONTI, “Article 81 EC and Public Policy”, (2002) 39(5) CMLR, p.1066. 99 The ECJ held in as early as 1966 that granting absolute territorial protection (“ATP”) to exclusive distributors by preventing exports of products by the distributor to other Member States or imports of these products from other Member States infringes Article 101(1). See Case-56 58/64, Etablissements Consten SARL and Grundig-Verkaufs-GmbH v Commission [1966] ECR 299. 100 Commission Regulation 330/2010 of 20 April 2010, on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices, [2010] OJ L102. 101 ZENGER & WALKER, supra note 76, p.196-197. 102 Ibidem, p.197. 103 Supra note 34. 104 Supra note 43, para 117. The specific circumstances concerned the regulation of prices in the pharmaceutical sector at national level. 105 Supra note 34, para 64.

19

been problematic106, it recognized the need to put competition law on parallel trade

restrictions more in line with competition economics.

The ECJ confirmed its formalistic object approach in two subsequent cases regarding

an export ban on decoding devices used for broadcasting football games in pubs, Football

Association Premier League107, and an absolute ban on sale of cosmetics and personal care

products via the internet, Pierre Fabre108. This line of cases demonstrates that classic object

restrictions are kept in the object box regardless of convincing economic theory stating that

they do not necessarily cause consumer harm. It appears that with regard to ATP internal

market objectives trump competition economics.109

2.1.2 Resale price maintenance (RPM)

“The dramatic evolution of US law in this area has fuelled the debate about the strict approach against […] RPM in the EU.”110 (Jones & Sufrin)

Imposing, directly or indirectly, fixed or minimum prices in a vertical agreement is

“blacklisted” under Article 4(a) of the VBER.111 However, economists argue that the

presumption of harm regarding RPM does not find support in competition economics.112

First, there is often a legitimate pro-competitive motive to introduce RPM. The aim is to

provide incentives for distributors to perform brand-specific investments, thereby increasing

non-price competition.113 Second, the fear of facilitating collusion by means of RPM is not

warranted where a manufacturer, facing strong inter-brand competition, applies the practices

unilaterally.114

In 2007, the US Supreme Court embraced economic thinking with respect to RPM. It

ruled in Leegin115 that imposing fixed or minimum prices in vertical agreements should

escape the illegality of a per se approach and fall under a rule of reason approach.

106 On the concerns regarding the abridged economic analysis approach, see Chapter 1.1.3. 107 Joined cases C-403/08 and C-429/08, Football Association Premier League Ltd and Others v QC Leisure and Others (not yet reported), judgment of 04.10.2011, paras 141-144. 108 C-439/09, Pierre Fabre Dermo-Cosmétique v President de l’Authorité de la Concurrence, [2010] OJ C24/49, para 47. 109 As Bishop and Walker put it, “the pursuit of market integration will sometimes clash with the pursuit of economic efficiency and consumer welfare”, see BISHOP & WALKER, The Economics of EC Competition Law, 3rd ed., Sweet & Maxwell, 2010, p.207. 110 JONES & SUFRIN, supra note 11, p.798. 111 Supra note 100. 112 ZENGER & WALKER, supra note 76, p.198-199. 113 Ibidem, p.198. 114 Ibidem, p.199. 115 Supra note 73.

20

Regardless of economic science and the recent developments in the US, the EU persists in

applying a presumption of harm towards RPM. This is not necessarily a bad policy decision.

However, in light of the recent Expedia judgment116 rendering appreciability arguments

under Article 101(1) impossible and the current concerns regarding the “death” of Article

101(3)117, the object approach to RPM in the EU is de facto almost identical as the

previously applied per se approach in the US, and therefore arguably out of tune with

competition economics.

2.2 Expanding the box beyond classic object restrictions

2.2.1 Reverse payment patent settlements (pay-for-delay)

" Today's decision is yet another warning that these so-called "pay for delay" agreements are illegal in the European Union "118 (J. Almunia)

2.2.1.1 Traditional position towards patent settlements: effects analysis Reverse payment patent settlements, also called “pay-for-delay”, “occur when a

patentee and an infringement defendant settle infringement litigation with an arrangement

under which the patentee pays the alleged infringer to stay out of the market for the period

covered by the settlement”.119 These settlements are common in the pharmaceutical industry

in the context of disputes between manufacturers of patent-protected drugs (the originators)

and manufacturers of generic drugs. While most IP settlements lead to the grant of a license,

where the main anti-competitive concern relates to the scope of the licensing agreement,

reverse payments settlements do not result in transfer of technology.120 Since, contrary to

most patent settlements, pay-for-delay agreements hinder potential early entry from

manufacturers of generics, a closer antitrust scrutiny is justified.

As regards the application of Article 101 to patent settlements, the Commission

initially held a very liberal view. In Bayer v Süllhöfer, the Commission regarded a no-

challenge clause as compatible with Article 101(1) when in the context of a genuine patent

settlement agreement, i.e. where “the existence of the industrial property right which is the

116 Supra note 17. 117 See infra Chapter 3.2. 118 Speech of J. ALMUNIA, “Fentanyl case”, Strasbourg, 10 December 2013. Press release is available at: http://europa.eu/rapid/press-release_SPEECH-13-1053_en.htm, 20 April 2014. 119 HOVENKAMP, “Antitrust and Patent Law Analysis of Pharmaceutical Reverse Patent Settlements”, 2011, p.1. Available at SSRN: http://ssrn.com/abstract=1741162, 20 April 2014. 120 Ibidem, p.1.

21

subject-matter of the dispute is genuinely in doubt”.121 The ECJ disagreed with the proposed

approach and held that the application of Article 101(1) to a no-challenge clause should not

be excluded, even if it the agreement is a bona fide patent settlement.122 Nevertheless, the

Court did not go as far as to state that such a restriction has an anti-competitive object. It is

striking how the Commission’s view on no-challenge clauses as part of patent settlements

changed from “compatible with Article 85(1)”123, when they involve the grant of a license, to

“by object”, when they prohibit entry of generic medicines and involve a (significant) value

transfer.124 This testifies of a drastic increase in scrutiny.

In its Final Report of the Pharmaceutical Sector Inquiry the Commission concluded

that, in order to assess the compatibility of a settlement agreement with EU competition law,

“an in-depth analysis of the individual agreement, taking into account the factual, economic

and legal background” is required.125 This can be taken to mean either that the Commission

adopts an effects-based approach or that the Commission can qualify patent settlements as

object restrictions relying on an extended contextual analysis similar to the one suggested by

the Court in Allianz Hungary126.

As regards the application of Article 101 to reverse payment patent settlements, the

Commission states in its new TTBER Guidelines127 that “pay-for-delay” settlements “are

based on a value transfer from one party in return for a limitation on the entry and/or

expansion on the market of the other party and may be caught by Article 101(1)”.128 The

choice of the word “may” means the patent settlements involving a transfer of value and

restriction of entry do not necessarily infringe Article 101(1).

To sum up, nothing in the early approach of the Commission or the ECJ, the more

recent Final Report on the Pharmaceutical Sector Inquiry or the new TTBER, provides a

warning sign that certain types of patent settlement agreements will be classified as object

restrictions.

121 Case 65/86, Bayer AG v Heinz Süllhöfer, [1988] ECR 05249, para 14. 122 Ibidem, paras 15-16. 123 Ibidem, para 14. 124 See infra notes 129,130 and 131. The main differences between Bayer v Süllhöfer and Lundbeck are the restriction of entry of generics, as opposed to the grant of a license, and the transfer of value. 125 European Commission Final Report of the Pharmaceutical Sector Inquiry of 8 July 2009, para 574. Available at: http://ec.europa.eu/competition/sectors/pharmaceuticals/inquiry/staff_working_paper_part1.pdf, 21 April 2014. 126 See supra note 15. 127 Commission Communication of 28 March 2014, Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to technology transfer agreements, O.J. C89/03, (hereinafter the “TTBER Guidelines”). 128 Ibidem, para 238.

22

2.2.1.2 Lundbeck et al.: object restrictions Two recent Commission infringement decisions, cases COMP/39.226 Lundbeck129

and COMP/39.685 J&J130, and one statement of objections, COMP/39.612 Servier131, put an

end to assessing all patent settlements on their anti-competitive effects. In all three cases the

Commission adopted an object approach and imposed hefty fines. Whether it makes sense

from a legal and economic point of view to qualify such practices as object restrictions, is

examined below.

Lundbeck involved a classic reverse payment patent settlement. The originator,

Lundbeck, paid several manufacturers of generic medicines not to challenge the validity of

its patent and to stay out of the market for citalopram, an antidepressant medicine, which

was also Lundbeck’s best selling drug.132 Although the patent for the citalopram molecule

had already expired, Lundbeck claimed exclusivity on account of its process patents.

Nevertheless, a few generic companies were preparing their entry and one of them had

already put its generic products on the market. Subsequently, the parties concluded a

settlement agreement, whereby the generic manufacturers agreed to delay their market entry

“in return for substantial payments and other inducements from Lundbeck amounting to tens

of millions of euros”. The Commission framed the practice as an object restriction. Since the

decision is not published yet, there is uncertainty as regards the decisive criteria in the case,

which convinced the Commission that there is no need to conduct an effects analysis. Based

on the information available in the press release, one can think of at least three propositions,

starting from the broadest approach towards object restrictions to the narrowest:

1. The Commission might defend the view that every patent settlement agreement

involving a value transfer and limiting entry has the object to restrict competition.

2. The Commission might reserve the notion of object restriction only for those patent

settlements where the value transfer is substantial.

129 Commission decision of 19.06.2013, not yet published. Press release is available at: http://europa.eu/rapid/press-release_IP-13-563_en.htm, 20 April 2014. 130 Commission decision of 10.12.2013, not yet published. Press release is available at: http://europa.eu/rapid/press-release_IP-13-1233_en.htm, 20 April 2014. 131 Commission statement of objections of 30.07.2012. Press release is available at: http://europa.eu/rapid/press-release_IP-12-835_en.htm, 20 April 2014. 132 The only source regarding the facts of the case is the press release, see supra note 129.

23

3. In addition to a substantial value transfer and limitation of entry, the Commission

might require that some evidence of anti-competitive intent be provided for a pay-

for-delay settlement to constitute an object restriction.133

While the first proposition completely disregards the possible legitimate objectives of

reverse payment settlements134, the third one seems to go against the idea that subjective

intent is not necessary for determining object restrictions135. Hence, the significant amount of

the value transfer is most likely to emerge as the pivotal criteria for distinguishing between

pay-for-delay settlements, which merit an effects analysis, and those that do not. Some

insider information presented by Mr Italianer in his speech at Fordham appears to confirm

the above assumption. He revealed that the Commission had evidence that “the value which

Lundbeck transferred took into consideration the profit or turnover the generic producer

expected if it had successfully entered the market”136. However, until the decision is

published, this assumption remains a mere speculation.

The J&J case was relatively more straightforward.137 Contrary to Lundbeck, it did not

concern a patent settlement because Johnson & Johnson’s patent on fentanyl, a pain-killer,

had already expired. Although the entry on the market was completely open for generic

manufacturers, the Dutch subsidiaries of Johnson&Johnson and Novartis, decided to

substitute competition for concluding a “co-promotion agreement”. Since its patent had

expired, the originator of the drug agreed to pay “monthly payments exceeding the profits

that the [generic manufacturer] expected to obtain”. In the absence of any other motive for

the value transfer, such as for example the incentive to settle and avoid litigation, the “co-

promotion agreement” was nothing less than a market-sharing agreement aimed at sharing

monopoly profits. In that respect, it was rightfully qualified as an object restriction by the

Commission.

Finally, in Servier the originator of perindopril, a cardio-vascular medicine, was

accused of both unilateral behavior, the acquisition of scarce competing technologies, and of

concluding patent settlement agreements hindering the entry of generic manufacturers.138 It

appears that the case concerned an unmeritorious patent that should not have been granted.

133 In Lundbeck, supra note 129, the Commission found internal documents referring to a "club" being formed and "a pile of $$$" to be shared among the participants. 134 See infra Chapter 2.2.1.3. 135 See supra note 33. 136 ITALIANER, supra note 30, p.9. 137 The only source regarding the facts of the case is the press release, supra note 130. 138 The only source regarding the facts of the case is the press release, supra note 131.

24

In Les Laboratoires Servier v Apotex Inc139, Lord Justice Jacob of the UK Court of Appeal

said that “it is the sort of patent which can give the patent system a bad name”. If the

invalidity of the patent was so obvious, then there could have been no other impulse for the

reverse payments than the restriction of competition.

2.2.1.3 Does pay-for-delay fit the object box? Having explained the previous position of the Commission to patent settlements and

the new approach adopted in its recent cases, we now turn to the question whether patent

settlements, involving a transfer value and a limitation of entry, should be qualified as object

restrictions, and if so, under what circumstances. As a reminder, Chapter 1 established that

where no past experience, based on economic theory of harm, demonstrates the significant

likelihood of harmful effects or where the restrictive clause is ancillary to the main non-

restrictive objective, the agreement should not be deprived of a full-fledged effects analysis.

It is submitted that pay-for-delay arrangements, except in some exceptional circumstances,

do not meet the appropriate standard for being regarded as object restrictions.

As regards the reliance on past experience, there are no existing precedents on

reverse payment settlements either from the Commission or the European Courts. It appears

that the Statement of Objections in Lundbeck relied to a great extent on BIDS140.141 The case

concerned agreements to pay processors of beef to exit the Irish beef market, purportedly in

order to reduce overcapacity. Although the two cases have some similarities, as they both

involve payments in return for limiting entry on or ensuring exit from the market, there is

one important distinction. Lundbeck was the holder of a patent which, if valid, grants the

right to exclude products from the market. This renders the use of BIDS as a precedent case

inappropriate.

As regards the robustness of the theory of harm, it is argued that pay-for-delay

arrangements are not always or almost always harmful. The main anti-competitive concern is

that “the patentee is using its monopoly profits to avoid the risk of patent invalidation or a

finding of noninfringement”.142 When such monopoly profits are shared with the generic

manufacturers in return for their staying out of the market, the agreement looks very much

139 [2008] EWCA Civ 445. 140 See supra note 25. 141 Presentation by SUBIOTTO, “Legal Assessment of Patent Settlement Agreements Containing “Reverse” Payments”, Fordham IP Conference, New York, 5 April 2013, slide 10. Available at: http://fordhamipconference.com/wp-content/uploads/2013/04/2013.subiotto.pharma.pdf, 20 April 2014 142 US Supreme Court, FTC v. Actavis, (2013), 570 U.S., p.17.

25

like a market-sharing arrangement. Nevertheless, a reverse payment patent settlement does

not necessarily lead to a collusive outcome.143 The uncertainty of the outcome is to a great

extent due to the probabilistic nature of patents144. There are at least three scenarios where

the exclusion in a pay-for-delay setting equals the exclusion in a plausible counterfactual

setting145:

1. The originator’s patent is valid, and therefore the generic manufacturer loses the

litigation.

2. The decision of the court is rendered only after the expiry of the patent, so that early

entry does not occur.

3. Generic manufacturer decides to discontinue the litigation.

In view of the various possible outcomes, applying a presumption of harm to reverse

payment settlements goes against economic theory. Hence, it is necessary to perform an

effects analysis. This involves establishing the appropriate counterfactual situation absent the

agreement and taking into account factors, such as the strength of the patent, the profits upon

entry for the generic manufacturer and its probability of winning the litigation.146

As regards the main objective of the agreement, it cannot be concluded with

sufficient certainty that the transfer of value necessarily aims at sharing monopoly profits. It

is argued that genuine fears of irreversible harm that might result from infringing entry by

generic firms147 makes it perfectly possible that the value transferred in return for limitation

of entry is ancillary to the objective of settling a bona fide dispute. Let us explore the

alternative to settlement for the patent holder – litigation. There are at least two downsides to

the choice of litigation: (i) it is often complex and costly and (ii) it might take years148. The

long duration of patent proceedings is a thorny issue for originators and becomes even more

apparent where a timely interim relief is not available.149 First, they will lose significant sales

due to the entry of generics. Second, since most EU Member States control prices of 143 OXERA ECONOMICS, "Bad medicine? An effects-based approach to 'pay-for-delay' agreements", October 2013. Available at: http://www.oxera.com/Latest-Thinking/Agenda/2013/Bad-medicine-An-effects-based-approach-to-pay.aspx, 21 April 2014. 144 Unlike other property rights, patents do not necessarily give the right to exclude rivals. Instead, they give the right to try to exclude rivals. In other words, there is always the possibility that the patent may be found invalid. See LEMLEY & SHAPIRO, “Probabilistic Patents”, 19(2) Journal of Economic Perspectives, Spring 2005, p.75-76. 145 OXERA ECONOMICS, supra note 143. 146 Ibidem. 147 SUBIOTTO, supra note 141, slide 21. 148 Patent litigation in the EU in the period between 2000 and 2007 lasted on average for 2.8 years, see Final Report of the Pharmaceutical Sector Inquiry, supra note 125, paras 635-636. 149 Interim relief in patent disputes in the EU was granted in less than 50% of the requests made by originator companies and its average duration was 18 months, see Final Report of the Pharmaceutical Sector Inquiry, supra note 125, paras 641-644.

26

medicines, the national health regulators of drug prices are likely to put the price down as a

result of the entry of generics. Once litigation is over, it would be a difficult and burdensome

task for the originator to convince the regulator to grant authorisation to sell again at the

previous higher price. Hence, the settlement alternative might be attractive even when the

originator strongly believes that he has a valid patent. These drawbacks of litigation testify

that entry of generics can cause originators to “suffer irreversible commercial losses” which

go much further than legal cost.150 Therefore, the value transfer, even if significant, should

not be taken to mean that the objective of the agreement could only be anti-competitive.

As a result of the lack of comparable precedents, the inconclusive harmful effects and

the plausible non-restrictive main objective, pay-for-delay arrangements merit a full-blown

effects analysis and not a treatment as object restrictions.151 Applying an (irrebuttable152)

presumption of harm to such agreements leads to an enforcement regime which is overly

prohibitive. It is likely to push parties to court litigation, thereby undermining the “public’s

“strong interest in settlement” of complex and expensive cases”153. Hence, it is of utmost

importance to establish the right balance between preserving the benefits of object

restrictions, namely increasing legal certainty and reducing enforcement costs, and

safeguarding incentives to settle bona fide disputes. So, under which circumstances would it

make sense to classify pay-for-delay arrangements as having an anticompetitive object?

It is submitted that there are three situations which justify the application of a

presumption of harm to reverse payment patent settlements:

1. “Beyond-scope” settlements: which go beyond the scope, including the duration, of

the contested patent.

2. “Sham” settlements: where it is obvious that the patent is invalid.

3. “Transfer-of-value” settlements, only where the value154 clearly exceeds the costs,

whether monetary or not, related to genuine litigation concerns. This is, for example,

a settlement where a generic manufacturer is compensated by the amount of its

expected profits if it is to successfully enter the market.155

150 See Final Report of the Pharmaceutical Sector Inquiry, supra note 125, para 640. 151 This was also the conclusion reached in Actavis, supra note 142, where the US Supreme Court adopted a “rule of reason” approach. 152 See infra Chapter 3. 153 Actavis, supra note 142, p.14. 154 The value is not necessarily in the form of a payment, but can also be in other forms, such as side deals. 155 The US Supreme Court in Actavis stated that “patentees sometimes pay a generic challenger a sum even larger than what the generic would gain in profits if it won” which may be taken to mean that “the patentee seeks to induce the generic challenger to abandon its claim with a share of its monopoly profits that would otherwise be lost in the competitive market”, supra note 142, p.15.

27

In all three circumstances, there is no other plausible motive for the settlement than sharing

the monopoly profits.

The Commission decisions in J&J and Servier appear to fall within the first and the

second category, respectively. Therefore the object approach to these two cases is justified.

Once the decision in Lundbeck is published, it will become apparent whether Lundbeck was

treated as an object restriction precisely because it fell within the third category described

above. If so, then fears of over-enforcement and chilling incentives to enter into bona fide

patent settlements in this field are exaggerated.

Whether the Commission will adopt an identical approach to pay-for-delay

arrangements as the one proposed above, namely an effects approach with three categories of

object restrictions, is of lesser importance. However, there is a pressing need to adopt a

policy that is in line with economic theory, easy to work with and clearly communicated to

the public, so that legal certainty in the sector can be restored.

2.2.2 Information sharing

“Almost under any circumstances information exchange could be good or bad.”156 (Kai-Uwe Kühn)

The enforcement policy towards information sharing has recently become more

intense. When exchange of information takes place within the framework of a broader cartel

arrangement, it forms part of a classic object restriction. When exchange of information

takes place within the framework of a cooperation agreement, an effects analysis should be

conducted. 157 Information sharing can also occur on a standalone basis: outside the

framework of a cartel or a cooperation agreement (“pure” information exchanges). It is the

Commission’s policy to treat “pure” information exchanges in certain circumstances as

object restrictions.

As illustrated by the above quote by Kai-Uwe Kühn, previous Chief Economist of

DG Comp, the effects of information sharing on the market are ambivalent. 158 Such

156 KUHN, “Designing Competition Policy towards Information Exchanges – Looking Beyond the Possibility Results” in OECD Report, Policy Roundtables: Information Exchanges Between Competitors under Competition Law (DAF/COMP(2010)37), p.416. 157 In that event information sharing is assessed within the framework of the cooperation agreement concerned, see Commission Communication of 14 January 2011, Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements [2011] O.J. C11/1 (“Horizontal Guidelines”), Sections 3-7. 158 KUHN, supra note 156.

28

uncertainty as to the positive or negative effects would normally warrant a case-by-case

analysis of effects instead of singling out a category of information exchanges presumed

harmful. However, in its Horizontal Guidelines the Commission states that “information

exchanges between competitors of individualised data regarding intended future prices or

quantities should therefore be considered a restriction of competition by object”.159 Most

economist 160 and lawyers161 agree to treating exchanges of future pricing or quantity

intentions as object restrictions. Such practices are likely to be employed as a means to

reaching a collusive outcome and are not likely to be justified by the existence of another

motive for the exchange than an anti-competitive one.162 Regrettably, it would appear that

classifying the exchange of future prices or quantities as object restrictions is simply an

example rather than the exception to the principle of effects analysis of information

sharing.163

In the famous T-Mobile judgment164 the ECJ produced some statements which, if

taken literally, eliminate any meaningful distinction between object and effect restrictions.

The case concerned a single meeting between mobile operators, where discussions between

competitors regarding future remuneration of mobile phone dealers took place. As regards

the notion of object restrictions, the Court appears to suggest that it is sufficient that a

practice “has the potential to have a negative impact on competition”.165 It could not

possibly have been the aim of the ECJ to establish a test of “potential” anti-competitive

effects in order to determine whether the restriction has the object of restricting competition.

Such an interpretation would leave no room for the analysis of restrictive effects under art

101(1). As Jones & Sufrin put it, this approach is “overly-expansive” and “worryingly

vague”.166 As regards information sharing, the Court ruled that it is “tainted with an anti-

159 Horizontal Guidelines, supra note 157, para 74. 160 BENNETT and COLLINS, “The Law and Economics of Information Sharing: the Good, the Bad and the Ugly”, (2010) 6(2) European Competition Journal, p.330; KUHN, supra note 156, p.422-423. 161 ORTEGA GONZALES, “Object analysis in information exchange among competitors”, (2012) 1(5) Global Antitrust Review, p.1-57, p.35. 162 The conclusion reached following the OECD policy roundtables was that whereas “private information exchanges and discussion about future intentions may be suitable for per se prohibitions”, “[o]ther types of exchanges should be considered on a case by case basis following a rule of reason type of analysis with a full evaluation of their effects and due deference to potential efficiencies”. See OECD Report, Policy Roundtables: Information Exchanges Between Competitors under Competition Law (DAF/COMP(2010)37), Executive Summary, p.12. 163 Horizontal Guidelines, supra note 157, para 72-73. 164 Case C-8/08, T-Mobile Netherlands BV v Raad van beestuur van de Nederlandse Mededingingsautoriteit [2009] ECR I-4529. 165 Ibidem, para 31. The reference to “potential” cannot simply be a mistake as the Court continues and explains that “the concerted practice must simply be capable in an individual case, having regard to the specific legal and economic context, of resulting in the prevention, restriction or distortion of competition within the common market”. 166 JONES & SUFRIN, supra note 11, p.213.

29

competitive object if the exchange is capable of removing uncertainties concerning the

intended conduct of the participating undertakings.”167 Such a broad test, based on the

capacity of the exchange to “remove uncertainties”, could catch many exchanges previously

dealt with under an effects approach. The unnecessarily168 wide view of the Court on when

on information sharing constitutes an object restriction might encourage the Commission and

other competition authorities to circumvent proof of effects in exchanges other than those

concerning future intentions.

167 T-Mobile, supra note 164, para 43 168 According to Meyring, the discussions between competitors in T-Mobile were ancillary to a price-fixing. Thus, there was no need for the Court to make such “bold statements” regarding “pure” information exchanges. See MEYRING, “T-Mobile: Further Confusion on Information Exchanges between Competitors”, (2010) 1(1) JECLAP, p.30-32.

30

3 The perverse effects: relying on presumptions Relying on presumptions in competition law is a pragmatic and sound approach.

However, it should not come at the cost of inaccurate results. Having demonstrated in

Chapter 2 that object restrictions are a broad concept, which may include practices that do

not always result in harm and may contain redeeming virtues, it is crucial that presumptions

remain rebuttable in order to avoid errors of over-enforcement. In other words, the defects of

the object box should not be worsened by overreliance on presumptions. It is submitted that

object restrictions give rise to the following presumptions: irrebuttable presumption of

appreciable anti-competitive effects (3.1) and a strong presumption of not satisfying the

conditions for Article 101(3) (3.2).

3.1 Irrebuttable presumption of appreciable anti-competitive effects

3.1.1 No need to show anti-competitive effects

“[F]or the purpose of applying Article [81(1)], there is no need to take account of the concrete effects of an agreement once it appears that it has as its object the prevention, restriction or distortion of competition.” 169 (ECJ in Consten & Grundig)

It is settled law that once an object restriction has been found to exist, it is

unnecessary to prove the existence of anti-competitive effects.170 This reflects the very

purpose of relying on a category of practices presumed harmful. The objective is to avoid

unnecessary, costly and time-consuming analysis of effects, where past experience and

economic science show that the practice always or almost always results in harm. A system

based on a category of practices, which are presumed anti-competitive, has many

advantages, is widely applied in other developed jurisdictions171 and is generally not

criticized172. The anti-competitive effects of an object restriction are not without any

relevance. They will be reflected in the amount of the fine and the award of damages.

169 Consten and Grundig, supra note 99, p.342. 170 Consten and Grundig, supra note 99; BIDS, supra note 25, para 16. 171 See supra Chapter 1.2.2. 172 JONES & SUFRIN, supra note 11, p.202.

31

3.1.2 No possibility to demonstrate absence of anti-competitive effects

“[T]he prohibition on ‘infringement by object’ may not be interpreted as meaning that an anti-competitive object gives rise merely to some kind of presumption of unlawfulness which may be rebutted, however, if in the specific case no negative consequences for the operation of the market can be demonstrated.” (AG Kokott)173

Not only is there no need for the Commission to establish anti-competitive effects,

but also there is no possibility for the undertaking to claim that the agreement did not (or

could not have) caused harm. In other words, the presumption of harm accompanying object

restrictions appears irrebuttable under Article 101(1).174

If the only rationale for object restrictions were to create a shortcut for establishing

the anti-competitive effects of certain practices that always or almost always cause harm, the

irrebuttability of the presumption of harm would seem counterintuitive. Provided that the

term “by object” were merely a quicker and less burdensome means of arriving at effects,

then why would an undertaking be prevented from demonstrating that the presumed harmful

effects did not materialize? This is where the logic of “risk offences” comes into play: an

object offence is punishable even when it did not cause any harm. As explained in Chapter

1.2.3, this rationale may make sense from a policy perspective when applied to hardcore

cartels, but not necessarily when applied to other restrictions, such as pay-for-delay, RPM

and ATP.

Odudu proposes to distinguish between “induction-based” presumptions, which

should remain rebuttable, and “intent-based” presumptions, which should be irrebuttable.175

Although sound from a theoretical point of view, this model would not be easy to work with,

as it relies on proof of intent, and so far finds no support in the case law. Hence, when an

object restriction is involved, the current state of the law precludes the undertaking from

showing absence of anti-competitive effects, even when the practice relates to non-cartel-like

behaviour where anti-competitive intent is not necessarily present.

173 AG Kokott in T-Mobile, supra note 16, para 45. 174 See T-Mobile, supra note 164, para 31. See KING, supra note 19, p.294; WHISH, supra note 27, p.117-118; NAGY, supra note 18, p.554. 175 ODUDU, The Boundaries of EC Competition Law: The Scope of Article 81, Oxford University Press, 2006, p.115.

32

3.1.3 No de minimis

“[A]n agreement that may affect trade between Member States and that has an anti-competitive object constitutes, by its nature and independently of any concrete effect that it may have, an appreciable restriction on competition. This is so regardless of the size and market share of the undertakings involved in the agreement”176 (ECJ in Expedia)

So far in this Chapter it has been established that an object restriction is conclusively

presumed to be harmful. The traditional approach to Article 101(1) requires that such harm

be appreciable and not negligible. In its De Minimis Notice177, the Commission quantifies

the appreciability requirement as a function of market shares. A “safe harbour” is established

where the parties to the agreement do not exceed certain market share thresholds.178 The

Commission excludes the benefit of the “safe harbour” to hardcore restrictions179, which are

typically the ones found in the object box. This exclusion does not lead to placing hardcore

restrictions in a “dangerous” harbour180. In other words, parties to an agreement that

contains a hardcore restriction can still claim before the courts that the restriction of

competition is not appreciable. This view finds support in the case-law. The European

Courts accepted on several occasions that object restrictions, in particular when in vertical

agreements, were de minimis. In the early Völk v Vervaecke, the ECJ ruled that an ATP

granted to an exclusive dealer “falls outside the prohibition in article [101] when it has only

an insignificant effect on the markets, taking into account the weak position which the

persons concerned have on the market of the product in question”.181 In the recent Pedro IV

Servicios, the ECJ stated that an RPM provision would only infringe Article 101 if it

“perceptibly” restricted competition.182 Hence, until recently it was settled case-law that de

minimis arguments were available to object restrictions.

A recent development seems to have set aside the traditional approach. Expedia

concerned a horizontal collaboration established by means of a joint venture between SNCF,

the French railway company, and Expedia, an online travel agency company. Expedia

benefited from a preferential treatment, namely a privileged access to SNCF’s online

booking system. The French competition authority found that this practice had as object to

176 Supra note 17, para 37. 177 Commission Notice of 3 September 1986, on agreements of minor importance which do not appreciably restrict competition under Article [101(1)] of the Treaty, [1986] O.J. C 231/2 (hereinafter the “De Minimis Notice”). 178 Ibidem, para 7. 179 Ibidem, para 11. 180 WHISH, supra note 27, p.141. 181 Case 5/69 Franz Völk v S.P.R.L. Ets J. Vervaecke [1969] ECR 295, para 6-7. 182 Case C-260/07 Pedro IV Servicios [2009] ECR I-2437, para 68.

33

exclude competition on a downstream market (i.e. the market for provision of online travel

services), and therefore infringed Article 101. The two parties to the agreement invoked a de

minimis argument, since their combined market share was below 10%. The ECJ, which was

referred the case, ruled that if the agreement had effect on inter-state trade and if it was

qualified as an object restrictions, it constituted automatically an appreciable restriction of

competition.183 It appears that this judgment establishes an irrebuttable presumption of

appreciable restriction of competition for object offences.184

Both lawyers and academics185 have criticized the judgment for being disconnected

with economic theory. Expedia would not have been problematic under the assumption that

object restrictions were to almost always result in harm and never produce benefits to

society. In other words, the irrebuttable presumption of appreciable anti-competitive effects

only makes sense when applied to agreements which have no possible redeeming virtues,

such as hardcore cartels. Indeed, why make a de minimis exception for an agreement which

can result in either neutral or negative impact on competition?186 However, as explained in

Chapter 2, the object box appears to contain at least a few restrictions which do have the

potential of being benign, such as pay-for-delay arrangements. In that respect, the Expedia

judgment only seems to exacerbate the problems related to the expansive interpretation of

object offences. Even if undertaken by parties with insignificant market power, such

practices are deemed anti-competitive. This reasoning is likely to dissuade small market

players from engaging in potentially welfare-enhancing practices. Hence, Expedia is a

regress towards formalism and goes against the aim of the modernisation process to put

competition law in line with economics.187

183 Expedia, supra note 176, para 37. 184 See Statement of the Commission in public consultations for the revision of the De Minimis Notice, available at http://ec.europa.eu/competition/consultations/2013_de_minimis_notice/index_en.html, 24 April 2014: “The Court has now established that the concept of a non-appreciable impact on competition (de minimis) does not apply when the agreement in question contains a so-called "by object restriction"”. 185 N. Petit nominated the judgment for the “worst antitrust development Oscar” in an article of 18 April 2013 in his blog Chillin’ Competition, available at: http://chillingcompetition.com/2013/04/18/law-firms-cartels/, 23 April 2014; See AKMAN, “The Court of Justice’s Expedia ruling undermines the economic approach by eliminating the ‘de minimis’ defence in object agreements”, available at: http://competitionpolicy.wordpress.com/2013/06/04/the-court-of-justices-expedia-ruling-undermines-the-economic-approach-by-eliminating-the-de-mimimis-defence-in-object-agreements/, 23 April 2014. 186 See NAGY, supra note 18. 187 One way to circumvent Expedia would be to apply a broader contextual (legal and economic background) approach to determining object restrictions, so as to account for very small market shares. This way the agreement will not be qualified as “by object” and will not bear the consequences of Expedia.

34

3.2 Presumption of not satisfying Article 101(3) conditions

“[T]he difficulty in this area is that a strong presumption has been created that such restraints will not satisfy the conditions of Article 101(3), and the modernisation process may have made it more difficult for those seeking to rebut that presumption.”188 (A. Jones)

Since Article 101(1) seems to establish an irrebuttable presumption of appreciable

harm for object restrictions, the only provision that separates the EU object approach from

the US per se illegality approach is Article 101(3). It is the exclusive forum under the EU

rules of competition for performing a balancing exercise between competitive harm and

redeeming virtues.189 The argument that efficiencies generated by the agreement offset the

restrictive effects is the only possible defence left for undertakings, once the object character

of the restriction has been proven.

In normative terms, the availability of Article 101(3) to object restriction is not

questioned. The General Court in Matra Hachette clearly states that “no anti-competitive

practice can exist which, whatever the extent of its effects on a given market, cannot be

exempted, provided that all the conditions laid down in Article 85(3) of the Treaty are

satisfied”.190 However, as argued below, it is the presumption of not satisfying the conditions

of Article 101(3), combined with the lack of sufficient guidance as to when the presumption

can be rebutted, that makes this provision de facto very difficult to rely on in practice in the

context of object restrictions.

First, Article 101(3) does not apply in the same way to object restrictions as it does to

effects restrictions. In its Article 101(3) Guidelines, the Commission states that “severe

restrictions of competition are unlikely to fulfil the conditions of Article [101(3)]”.191 It

refers to the hardcore restrictions found in different block exemption regulations, which are

typically the classic vertical and horizontal agreements in the object box. Such restrictions

are, as stated in the Article 101(3) Guidelines, unlikely to satisfy any of the first three of the

four cumulative conditions. 192 No such presumption applies with respect to effects

restrictions. In accordance with the rules on burden of proof stipulated in Regulation 1/2003,

it is up to the alleged infringer to demonstrate that the conditions for Article 101(3) are

188 JONES, supra note 13, p.669. 189 ITALIANER, supra note 30, p.7. 190 Case T-17/93, Matra Hachette, [1994] ECR II-595, para 85. 191 Supra note 25, para 46. 192 The four cumulative conditions of Article 101(3) are as follows: (i) efficiency gains; (ii) fair share for consumers; (iii) indispensability of the restrictions; (iv) no elimination of competition.

35

fulfilled.193 Given the existence of this strong presumption, claiming an exemption for object

restrictions looks like an “uphill battle”.

Second, in the absence of clear guidance as regards the circumstances in which the

presumption can be rebutted, companies are more inclined to abandon their potentially

welfare-enhancing business. The unattractive alternative is to fight an “uphill battle”, which

is very likely to end in damaging consequences, namely the imposition of a large fine.

Reports of the “slow death of Article 101(3)” since the advent of the modernisation process

have become quite common amongst lawyers.194 In addition, academics have recognized the

concerns related to the “paucity of cases providing guidance” as well.195 Such disquietude

did not exist under the “old regime”, which required notification of potentially restrictive

agreements.196 The Commission, having the exclusive competence to apply Article 101(3),

exempted many agreements and provided useful guidance as to why the specific restraints

satisfied the four cumulative criteria.

The 2004 modernisation wave brought about two main changes: it abolished the

notifications system and gave Article 101(3) direct effect197. The outcry related to the

absence of guidance to private parties is the result of three main post-modernisation

developments. Firstly, the post-notification decisional practice does not contribute to the

development of Article 101(3). Although Article 10 of Regulation 1/2003 provides the

Commission with the possibility to issue non-infringement, or finding of inapplicability,

decisions198, no such decisions have been issued since 2004. As a consequence, post-

notification Commission decisions do not shed light on when agreements merit exemption

under Article 101(3). In addition, the increasing preference for commitment decisions raises

concerns amongst academics and lawyers about the lack of guidance on novel and non-

193 Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, [2003] OJ L 1 (hereinafter “Regulation 1/2003”), Article 2. 194 See LAMADRID, “The Slow Death of Article 101(3)” in Chillin’ Competition blog, 28.11.2011, available at http://chillingcompetition.com/2011/10/28/the-slow-death-of-article-1013/, 24 April 2014; ADKINS, BEIGHTON, “An Object of Interest”, (2014) Competition Law Insight. Available at http://www.wragge.com/published_articles_11035.asp, 21 April 2014; Contra ITALIANER, supra note, p.8: “[R]eports of the death of Article 101(3) for object restrictions have been greatly exaggerated.” 195 JONES, supra note 13, p.668-670. 196 Council Regulation 17 [1959–62] OJ Spec Ed 87, Article 9(1). 197 NCAs and national judges could from then on apply Article 101(3), see Regulation 1/2003, supra note 193, Articles 5 and 6. 198 Recital 14 of the Preamble to Regulation 1/2003 further explains that the insertion of Article 10 is “with a view to clarifying the law and ensuring its consistent application throughout the [Union], in particular with regard to new types of agreements or practices that have not been settled in the existing case-law and administrative practice”.

36

obvious infringements.199 Secondly, the recent Tele2Polska judgment of the ECJ, prohibiting

NCAs from issuing non-infringement decisions200, may reduce the prospects of useful

guidance coming from Member States. Finally, the Article 101(3) Guidelines are denounced

as being too general and complex.201 They were adopted in 2004 with a view to providing

private parties with guidance on how to perform a self-assessment exercise and to ensuring

that national courts and NCAs apply the provision in a uniform way. Albeit not completely

unhelpful, they cannot fully compensate for the industry-specific analysis performed by the

Commission under the “old regime”.202 Furthermore, since they are based on the pre-

modernisation decisional practice and case-law, the Article 101(3) Guidelines do not contain

any guidance on how to apply this provision to some of the new types of object restrictions,

such as pay-for-delay arrangements.

As opposed to Article 101(1), which does not seem to allow undertakings any

possibility of defence once an object restriction is established, Article 101(3) is not a “fully

closed door”. However, in practice, the strong presumption of not complying with the

required criteria coupled with the lack of guidance on how to rebut the presumption in

specific cases, dissuades parties from entering into potentially welfare-enhancing

agreements.203 The more the object box expands, so as to include agreements with the

capacity of having redeeming virtue, the greater the need for a revival of Article 101(3).

199 However, commitment decisions may serve as a source of informal guidance. See Commission commitment decision of 23.05.2013, Case COMP/AT.39595, Continental/United/Lufthansa/Air Canada, [2013] C(2013) 2836 final (hereinafter “Star Alliance”), paras 56-86, where the Commission explained how it applies Article 101(3) to certain forms of cooperation between airline alliances. 200 Case C-375/09, Prezes Urz edu Ochrony Konkurencji i Konsumento w v Tele2 Polska sp. z o.o., [2011] ECR I-3055, para 30. While an NCA may find that “there are no grounds for action on their part”, only the Commission can find that Articles 101 and 102 are inapplicable to a particular practice. It seems that after Tele2Poslka the power of NCAs to apply Article 101(3) is limited to infringements decisions. See MACGREGOR, GECIC, “EU Antitrust Proceedings and National Competition Authorities: A Leap in the Wrong Direction”, [2012] International Trade Law & Regulation, p.1-9. 201 See GERARD, supra note 13, p.36. 202 Ibidem. 203 ADKINS, BEIGHTON, “An Object of Interest”, (2014) Competition Law Insight. Available at http://www.wragge.com/published_articles_11035.asp, 21 April 2014

37

Conclusion This paper has demonstrated that since the advent of the modernisation EU

competition law has broadened the concept of object restrictions in two ways.204 On the one

hand, after Allianz Hungária it seems acceptable to classify practices, typically falling

outside of the object box, as object restrictions by means of an abridged competitive

assessment. The far-reaching claws of this assessment risk reducing legal certainty and

blurring the distinction between effect and object. On the other hand, new types of

infringements, such as pay-for-delay arrangements and certain types of “pure” information

exchanges, have enlarged the size of the object box. The addition of new practices to the list

of object restrictions is not by and of itself objectionable, since the object category is a not

static, but a dynamic concept. However, every new inclusion should be accompanied by a

“bulletproof” theory of harm. 205 Furthermore, it is recommended that, the first cases before

the insertion of a practice into the object box be given a full-blown economic analysis.

To avoid errors of over-enforcement, the broad view of the EU system towards a

finding of an object restriction should be accompanied by a broad view of the available

defences under both Article 101(1) and 101(3). This paper has argued that EU competition

law appears to evolve in the opposite direction. The expansion of the object box is combined

with a significant limitation of the possibilities to demonstrate that the practices are legal.

The perverse effects of expanding the object box are that, in addition to the application of a

presumption of harm, which is not on all occasions warranted on grounds of economics, the

defendant is precluded from arguing that the expected restriction of competition did not

materialize or that, if it did, it was negligible. What is more, the defendant faces a strong

presumption, that the agreement does not merit exemption under Article 101(3). As a result

of the absence of sufficient guidance on how to rebut the presumption, parties often prefer

giving up their potentially welfare-enhancing practices to risking a hefty fine.

Although the EU “object box” is clearly larger than the US “per se box”, as the

former includes practices, such as RPM, ATP, some information exchanges and some pay-

for-delay arrangements, there is not much that separates the consequences stemming from an

object restriction from the ones stemming from a per se infringement. The EU combination

of a broad object box with limited defence options is more likely to condemn an “innocent”

man than to save a “guilty” one.

204 For a schematic overview, see Annex. 205 It was argued in this paper that, on account of economic theory, it is justified to include in the object box exchanges related to future intentions and the three examples of pay-for-delay arrangements given on page 26.

38

Bibliography BOOKS

BISHOP & WALKER, The Economics of EC Competition Law, 3rd ed., Sweet & Maxwell, 2010

FAULL, & NIKPAY, The EU Law of Competition, 3rd ed., Oxford University Press, Oxford, 2014

JONES & SUFRIN, EU Competition Law, 5th ed., Oxford University Press, 2014 ODUDU, The Boundaries of EC Competition Law: The Scope of Article 81, Oxford

University Press, 2006 PELLETIER, La caducité des actes juridiques en droit privé français, Paris, Harmattan, 2004

PETIT, Droit européen de la concurrence, Paris, Montchrestien, 2013 WHISH, Competition Law, 7th ed., Oxford University Press, 2012

ARTICLES

ANDREANGELI, “From Mobile Phones to Cattle: How the Court of Justice Is Reframing the Approach to Article 101 (Formerly 81 EC Treaty) of the EU Treaty”, (2011) 34(2) World Competition, p.215–243

AKMAN, “The Court of Justice’s Expedia Ruling Undermines the Economic Approach by Eliminating the ‘De Minimis’ Defence in Object Agreements”. Available at: http://competitionpolicy.wordpress.com/2013/06/04/the-court-of-justices-expedia-ruling-undermines-the-economic-approach-by-eliminating-the-de-mimimis-defence-in-object-agreements/, 23 April 2014

BAILEY, “Restrictions of Competition by Object under Article 101 TFEU”, (2012) 49 CMLR, p.559-600

BENNETT and COLLINS, “The Law and Economics of Information Sharing: the Good, the Bad and the Ugly”, (2010) 6(2) European Competition Journal, p.311-337

COLOMO, “Market Failures, Transaction Costs and Article 101(1) TFEU Case Law”, (2012) 37(5) E.L.Rev., 541-562

DICKEY & RUBINFELD, “Would the Per Se Illegal Treatment of Reverse Payments Settlements Inhibit Generic Drug Investment”, (2012) 8(3) Journal of Competition Law & Economics, 615-625

GERARD, “The Effects-Based Approach Under Article 101 TFEU and its Paradoxes: Modernisation at War with Itself?” in BOURGEOIS and WAELBROECK (eds.), Ten Years of Effects-Based Approach in EU Competition Law, Bruylant, 2012, p.17-41

GOYDER, “Cet Obscur Objet: Object Restrictions in Vertical Agreements” (2011) 2 JECLP 327

HARRISON, “The Allianz Hungária case - The ECJ's judgment could have ugly consequences”, (2013) 12(6) Competition Law Insight, p.10-12

HOVENKAMP, “Antitrust and Patent Law Analysis of Pharmaceutical Reverse Patent Settlements”, 2011, p.1-20. Available at SSRN: http://ssrn.com/abstract=1741162, 20 April 2014

39

JONES, “Left Behind by Modernisation? Restrictions by Object under Article 101(1)”, (2010) 6(3) European Competition Journal, p.649-676

KING, “The Object Box: Law, Policy or Myth?”, (2011) 7(2) European Competition Journal, p.269-296

KOMOSSA, “Allianz Hungária”, (2013) 4(5) JECLP, p.417-419 LEMLEY & SHAPIRO, “Probabilistic Patents”, 19(2) Journal of Economic Perspectives,

Spring 2005, p.75-98 MACGREGOR, GECIC, “EU Antitrust Proceedings and National Competition Authorities: A

Leap in the Wrong Direction”, [2012] International Trade Law & Regulation, p.1-9 MAHTANI, "Thinking Outside the Object Box: An EU and UK Perspective", (2012) 8(1)

European Competition Journal, p.1-39 MEYRING, “T-Mobile: Further Confusion on Information Exchanges between Competitors”,

(2010) 1(1) JECLAP, p.30-32 MONTI, “Article 81 EC and Public Policy”, (2002) 39(5) CMLR, p.1057-1099

NAGY, “The Distinction between Anti-competitive Object and Effect in Allianz: The End of Coherence in Competition Analysis?”, (2013) 36(4) World Competition, p.541-564

ODUDU, "Restrictions of Competition by Object - What's the Beef?", (2009) 9(1) Competition Law Journal, p.11-17

ODUDU, “Interpreting Article 81(1): Object as subjective intention”, (2001) 26(1) E.L.Rev., p.60-75

ORTEGA GONZALES, “Object analysis in information exchange among competitors”, (2012) 1(5) Global Antitrust Review, p.1-57

OXERA ECONOMICS, "Bad medicine? An effects-based approach to 'pay-for-delay' agreements", October 2013. Available at: http://www.oxera.com/Latest-Thinking/Agenda/2013/Bad-medicine-An-effects-based-approach-to-pay.aspx, 21 April 2014

RO LLER, L-H & DE LA MANO, M., “The Impact of the New Substantive Test in European merger Control” (2006) 2(1) European Competition Journal, p.9-28.

VEDDER, “Allianz and the Object-Effect Dichotomy in Article 101(1) TFEU: A Practical Solution Meets Not So Practical Competition Law”, European Law Blog, 8 April 2013. Available at: http://europeanlawblog.eu/?p=2147#sthash.6FYzqXCT.dpuf,

ZENGER & WALKER, “Theories of Harm in European Competition Law: A Progress Report” in BOURGEOIS and WAELBROECK (eds.), Ten Years of Effects-Based Approach in EU Competition Law, Bruylant, 2012, p.185-209

40

CASE-LAW OF THE ECJ Case C-32/11, Allianz Hungária Biztosító Zrt and Others v Gazdasági Versenyhivatal (not

yet reported), judgment of 14.03.2013 Case C-226/11, Expedia Inc. v. Autorité de la Concurrence (not yet reported), judgment of

13.12.2012 Case C-457/10 P, AstraZeneca AB and AstraZeneca plc v. European Commission (not yet

reported), judgment of 6.12.2012 Case C-209/10, Post Danmark A/S v Konkurrencera det (not yet reported), judgment of

27.03.2012 C-439/09, Pierre Fabre Dermo-Cosmétique v President de l’Authorité de la Concurrence,

[2010] OJ C24/49 Case C-375/09, Prezes Urz edu Ochrony Konkurencji i Konsumento w v Tele2 Polska sp. z

o.o., [2011] ECR I-3055 Joined cases C-403/08 and C-429/08, Football Association Premier League Ltd and Others v

QC Leisure and Others (not yet reported), judgment of 04.10.2011 Case C-8/08, T-Mobile Netherlands BV v Raad van beestuur van de Nederlandse

Mededingingsautoriteit [2009] ECR I-4529 Case C-260/07 Pedro IV Servicios [2009] ECR I-2437

Case C-209/07, Competition Authority v. Beef Industry Development Society and Barry Brothers, [2008] ECR I-8637

Joined Cases C-501/06, C-513/06, C-515/06 and C-519/06, GlaxoSmithKline v. Commission, [2009] ECR. I-9291

Case C-519/04 P, David Meca-Medina and Igor Majcen v Commission of the European Communities, [2006] ECR I-06991

Case C-7/95P, John Deere Ltd v. Commission, [1998] ECR I-3111 Case C-234/89, Delimitis [1991] ECR I-935

Case 27/87, Erauw-Jacquéry v. La Hesbignonne, [1988] ECR 1919 Case 65/86, Bayer AG v Heinz Süllhöfer, [1988] ECR 05249

Case 262/81, Coditel v Ciné Vog Films, [1982] ECR 3381 Case 5/69 Franz Völk v S.P.R.L. Ets J. Vervaecke [1969] ECR 295

Case C-56/65 Société Technique Minière (L.T.M.) v Maschinenbau Ulm GmbH (M.B.U.) [1966] ECR 235

Case-56 58/64, Etablissements Consten SARL and Grundig-Verkaufs-GmbH v Commission [1966] ECR 299

41

CASE-LAW OF THE GENERAL COURT Case T-199/08, Ziegler SA v European Commission, [2011] E.C.R. II-03507

Joined Cases T-49-51/02, Brasserie Nationale v. Commission, [2005] ECR II-3033 Case T-168/01, GlaxoSmithKline Services v Commission, [2006] ECR II-2969

Joined Cases T-374, 375, 384 & 388/94, European Night Services v. Commission, [1998] ECR II-3141

Case T-17/93, Matra Hachette, [1994] ECR II-595

CASE-LAW OF THE US SUPREME COURT US Supreme Court, FTC v. Actavis, (2013) 570 U.S

US Supreme Court, Leegin Creative Leather Products Inc v PSKS, Inc, (2007) 551 US 877 US Supreme Court, Jacobellis v. Ohio, (1964) 378 U.S. 184

US Supreme Court, Northern Pacific Railway Co. v. United States, (1958) 356 U.S. 5

OTHER CASE-LAW ECtHR, Menarini Diagnostics S.R.L. v. Italy, [2011] No 43509/08

UK Court of Appeal, Les Laboratoires Servier v Apotex Inc, [2008] EWCA Civ 445

COMMISSION DECISIONS Commission Decision of 10.12.2013, COMP/39.685, J&J, not yet published. Press release is

available at: http://europa.eu/rapid/press-release_IP-13-1233_en.htm, 20 April 2014 Commission Decision of 19.06.2013, COMP/39.226, Lundbeck, not yet published. Press

release is available at: http://europa.eu/rapid/press-release_IP-13-563_en.htm, 20 April 2014

Commission Commitment Decision of 23.05.2013, COMP/AT.39595, Continental/United/Lufthansa/Air Canada, [2013] C(2013) 2836 final

Commission Statement of Objections of 30.07.2012, COMP/39.612, Servier. Press release is available at: http://europa.eu/rapid/press-release_IP-12-835_en.htm, 20 April 2014

Commission Decision of 13.05.2009, COMP/37.990, Intel, OJ C227/13

OPINIONS OF ADVOCATES-GENERAL Opinion of AG Wahl in Case C-67/13 P, Groupement de cartes bancaires (CB) contre

Commission européenne, delivered on 27 March 2014 Opinion of AG Cruz Villalón in Case C-32/11, Allianz Hungária, delivered on 25 October

2012 Opinion of AG Kokott in Case C-226/11, Expedia Inc. v Autorité de la concurrence and

Others, delivered on 6 September 2012 Opinion of AG Sharpston in Case C-272/09 P, KME Germany AG v. Comission, delivered

on 10 February 2011

42

Opinion of Advocate General Trstenjak in Case C-209/07, Competition Authority v Beef Industry Development Society Ltd and Barry Brothers (Carrigmore) Meats Ltd, delivered on 10 January 2009

Opinion of AG Kokott in Case C-8/08, T-Mobile, delivered on 19 February 2008

EU SECONDARY LEGISLATION

Commission Regulation 330/2010, on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices, [2010] O.J. L102

Council Regulation (EC) No 139/2004 of 20 January 2004, on the control of concentrations between undertakings, [2004] O.J. L 24

Council Regulation (EC) No 1/2003 of 16 December 2002, on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, [2003] O.J. L 1

Council Regulation 17 of 21 February 1962, [1959–62] O.J. Spec Ed 87

Commission Communication of 28 March 2014, Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to technology transfer agreements, O.J. C89/03

Commission Communication of 14 January 2011, Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements [2011] O.J. C11/1

Commission Communication of 24 February 2009, Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings, O.J. C 45

Commission Communication of 27 April 2004, regarding Guidelines on the Application of Article 81(3), [2004] O.J. C101/97

Commission Notice of 3 September 1986, on agreements of minor importance which do not appreciably restrict competition under Article [101(1)] of the Treaty, [1986] O.J. C 231/2

White Paper of 28 April 1999, on Modernisation of the Rules Implementing Articles 85 and 86 of the EC Treaty, Commission Programme No 99/027, COM (99) 101 final. Available at: http://europa.eu/documents/comm/white_papers/pdf/com99_101_en.pdf, 23 April 2014

OTHER SOURCES OECD Report, Fighting Hard-Core Cartels: Harm, Effective Sanctions and Leniency

Programmes, OECD Publications, Paris, 2002 OECD Report, Policy Roundtables: Information Exchanges Between Competitors under

Competition Law DAF/COMP(2010)37, 2010 KUHN, “Designing Competition Policy towards Information Exchanges – Looking Beyond

the Possibility Results” in OECD, Policy Roundtables: Information Exchanges Between Competitors under Competition Law DAF/COMP(2010)37, 2010

43

European Commission Final Report of the Pharmaceutical Sector Inquiry of 8 July 2009. Available at: http://ec.europa.eu/competition/sectors/pharmaceuticals/inquiry/staff_working_paper_part1.pdf, 21 April 2014

Speech by ITALIANER, “Competitor Agreements under EU Competition Law”, 40th Annual Conference on International Antitrust Law and Policy at Fordham, New York, 26 September 2013

Speech by ALMUNIA, “Fentanyl case”, Strasbourg, 10 December 2013. Press release is available at: http://europa.eu/rapid/press-release_SPEECH-13-1053_en.htm, 20 April 2014

Presentation by SUBIOTTO, “Legal Assessment of Patent Settlement Agreements Containing “Reverse” Payments”, Fordham IP Conference, New York, 5 April 2013. Available at: http://fordhamipconference.com/wp-content/uploads/2013/04/2013.subiotto.pharma.pdf, 20 April 2014

ANNEX: Expanding the Object Box

Classic object restrictions

Effects Analysis

Per se illegality

Information sharing removing

uncertainties

Reverse patent settlements with transfer of value

Contextual analysis approach

RPM

ATP

New infringements

Abridged competitive analysis

About Mayer Brown Mayer Brown is a global legal services organisation advising clients across the Americas, Asia and Europe. Our presence in the world’s leading markets enables us to offer clients access to local market knowledge combined with global reach.

We are noted for our commitment to client service and our ability to assist clients with their most complex and demanding legal and business challenges worldwide. We serve many of the world’s largest companies, including a significant proportion of the Fortune 100, FTSE 100, DAX and Hang Seng Index companies and more than half of the world’s largest banks. We provide legal services in areas such as banking and finance; corporate and securities; litigation and dispute resolution; antitrust and competition; US Supreme Court and appellate matters; employment and benefits; environmental; financial services regulatory & enforcement; government and global trade; intellectual property; real estate; tax; restructuring, bankruptcy and insolvency; and wealth management.

OFFICE LOCATIONS

AMERICAS• Charlotte • Chicago • Houston • Los Angeles • New York • Palo Alto • Washington DC

ASIA• Bangkok • Beijing • Guangzhou • Hanoi • Ho Chi Minh City • Hong Kong • Shanghai • Singapore

EUROPE• Brussels • Düsseldorf • Frankfurt • London • Paris

TAUIL & CHEQUER ADVOGADOSin association with Mayer Brown LLP• São Paulo • Rio de Janeiro

Please visit www.mayerbrown.com for comprehensive contact information for all Mayer Brown offices.

Mayer Brown is a global legal services provider comprising legal practices that are separate entities (the “Mayer Brown Practices”). The Mayer Brown Practices are: Mayer Brown LLP and Mayer Brown Europe–Brussels LLP, both limited liability partnerships established in Illinois USA; Mayer Brown International LLP, a limited liability partnership incorporated in England and Wales (authorised and regulated by the Solicitors Regulation Authority and registered in England and Wales number OC 303359); Mayer Brown, a SELAS established in France; Mayer Brown JSM, a Hong Kong partnership and its associated entities in Asia; and Tauil & Chequer Advogados, a Brazilian law partnership with which Mayer Brown is associated. “Mayer Brown” and the Mayer Brown logo are the trademarks of the Mayer Brown Practices in their respective jurisdictions.

This publication provides information and comments on legal issues and developments of interest to our clients and friends. The foregoing is not a comprehensive treatment of the subject matter covered and is not intended to provide legal advice. Readers should seek legal advice before taking any action with respect to the matters discussed herein.

© 2014. The Mayer Brown Practices. All rights reserved.

June 2014